Effect of paternal body mass index on maternal and child-health outcomes of singletons after frozen-thawed embryo transfer cycles: a retrospective study

Abstract

The objective was to analyze the effect of paternal body mass index (BMI) on maternal and child-health outcomes of singletons after frozen-thawed embryo transfer (FET) cycles. A retrospective cohort study was conducted between January 2019 and December 2021. Pregnancy, perinatal complications and neonatal outcomes were compared among different paternal BMI. Multivariate logistic regression was performed to evaluate the relationship between different paternal BMI and pregnancy, obstetric and neonatal outcomes. The paternal normal group was more likely to suffer from gestational hypertension than the paternal obesity group (3.59% vs. 2.42%), and paternal underweight group was more likely to suffer from preeclampsia than the other three groups (11.63% vs. 4.43%, 7.57%, 4.03%). Birthweight among infants in the paternal overweight categories was significantly higher than infants in the paternal normal weight categories. The rate of foetal macrosomia was higher among infants in the paternal overweight (12.36%) category, while lower among infants in the paternal underweight categories (2.33%). The incidence of macrosomia in the paternal overweight categories (aOR 1.527, 95% CI 1.078–2.163) was significantly higher than those normal controls after adjustment for known confounding factors. The rates of LGA babies were higher in the paternal overweight category (aOR 1.260, 95% CI 1.001–1.587) compared with those in the paternal normal weight category, before and after adjustment. The results suggest that parental pre-pregnancy overweight or obesity has an adverse effect on the perinatal complications and neonatal outcomes.

Keywords:

• Paternal
• body mass index
• assisted reproductive technology
• frozen embryo transfer
• pregnancy complications
• neonatal outcomes

Introduction

As an important issue in healthcare, obesity is a kind of global epidemic and weight management point before and during pregnancy (Broughton & Moley, 2017). The prevalence of obesity has grown rapidly worldwide. According to the World Health Organization statistics, 64.1% men were estimated to be overweight and obese in the United States, 3.2% higher than the rate among women (Vardell, 2020). Overweight and obesity are defined as abnormal or excessive fat accumulation that threatens the health of the individual. For adults, the World Health Organization (WHO) defines overweight and obesity as a BMI ≥25 and ≥30, respectively. BMI provides the most useful population-level measure of overweight and obesity (Caballero, 2019). The BMIs of Asian populations are generally lower than non-Asian populations (World Health Organization Expert Consultation, 2004). Common overweight and obesity-related health problems include cardiovascular disease (CVD), hypertension, type 2 diabetes (T2D), hyperlipidaemia, stroke, certain cancers, sleep apnoea, liver and gall bladder disease, osteoarthritis, and gynaecological issues (Nestle & Jacobson, 2000). Over the past decades, attention has been continuously directed towards the effect of obesity on fertility.

The influence of maternal BMI and obesity on pregnancy and child health has been extensively researched, with systematic reviews finding increased rates of pregnancy/gestation complications including gestational diabetes, pre-eclampsia, hypertension, depression, instrumental and caesarean birth, preterm birth, surgical site infection, as well as neonatal complications including perinatal death, macrosomia, foetal defects and congenital anomalies in women of increasing BMI (Marchi et al., 2015).

The impact of paternal BMI is much less known. Elevated body mass index (BMI) may have adverse effects on male fecundity. Paternal BMI has been associated with a negative impact on embryo quality and in vitro fertilization (IVF) outcomes (Liu et al., 2017; Mushtaq et al., 2018). However, the results in these studies are not consistent. A meta-analysis has found that increased paternal BMI is associated with a higher risk of small-for-gestational age (SGA) and macrosomia (Campbell & McPherson, 2019). Nevertheless, a number of studies observed that paternal BMI had no association with birth weight, large-for-gestational age (LGA) and SGA (Magnus et al., 2001; Takagi et al., 2019). Up to date, the association between elevated male BMI and the clinical results after ART remains controversial.

Since abnormal birthweight increases the risk of complications during the foetal and neonatal periods and is also a strong predictor for subsequent adulthood metabolic and cardiovascular disease, it is crucial to identify whether abnormal paternal BMI, especially overweight and obesity, are risk factors for adverse neonatal outcomes. The frozen-thawed embryo transfer (FET) cycle provides an excellent model for studying the influence of paternal factors on neonatal outcomes after IVF/ICSI treatment, as it allows the possibility of adverse foetal growth arising from a hyper-oestrogenic milieu in fresh ET cycles to be ruled out (Imudia et al., 2012; Pereira et al., 2017). As a result, the purpose of the present study was to clarify the relationship between paternal BMI and maternal and child-health outcomes of singletons using FET cycles.

Materials and methods

Study design and participants

A retrospective study was performed in the Department of Assisted Reproduction at the Women’s Hospital of Nanjing Medical University. This study was approved by the Clinical Research Ethics Committee of the Nanjing Maternity and Child Health Care Hospital (2021KY-111). Women who underwent FET during the period from January 2019 to December 2021 were enrolled. Women who had normal BMI (18.5 kg/m² ≤ BMI < 25 kg/m²) and a live singleton birth were included. The inclusion criteria were as follows: (1) need for assisted reproduction therapy (ART) with IVF or intracytoplasmic sperm injection (ICSI), (2) the age ≤42 years during oocyte retrieval, (3) one or more high-quality embryos had been transferred on the day of ET. The exclusion criteria were as follows: (1) maternal age >45 years, (2) couples with severe complication before pregnancy, such as diabetes, hypertension, heart or liver disease, (3) couples with a history of smoking or drinking, (4) couples undergoing preimplantation genetic testing, and (5) data were incomplete or incorrect in the database. In the case of patients who had more than one delivery during the study period, only the first pregnancies were included for analysis. A history of smoking was defined as a patient who smoked 1 or more cigarettes per day for at least six months previously (World Health Organization, 1998). A history of drinking was defined as 60 or more grams of pure alcohol on at least one single occasion in the past seven days (World Health Organization, 2014). A total of 1,609 women were included for the final analysis.

According to the classification and evaluation criteria of the World Health Organization, all infants were divided into four categories based on paternal BMI: paternal underweight (BMI < 18.50 kg/m²), paternal normal weight (18.50 kg/m²≤ BMI < 25 kg/m²), paternal overweight (25 kg/m² ≤ BMI < 30 kg/m²) and paternal obesity (BMI ≥ 30 kg/m²). BMI was measured by a trained nurse at the start of treatment initiation. Paternal normal weight was used as a control group.

Data collection

We searched the electronic medical database to retrieve the data of maternal characteristics and treatments in ART, including female age, BMI, cause of infertility, duration of infertility, smoking history, parity, insemination methods, stimulation protocol in FET cycle, number of embryos transferred, endometrial thickness. Pregnancy and perinatal outcomes, including pregnancy-induced hypertension (HDP), gestational diabetes or pregestational diabetes mellitus, newborn gender, mode of delivery, gestational age, birth weight and Apgar score, were also obtained.

Endometrial preparation and frozen-thawed embryos

Details on endometrial preparation procedures have been described in our previous study (X. Li et al., 2022). Briefly, patients with regular ovulatory cycles were treated with modified natural cycles (MNC), in which ultrasound monitoring was started at 10–12^th day of the menstrual cycle. For women with irregular menses, endometrial preparation was carried out in either an ovulation induction (OI) cycle or an artificial cycle (AC).

Embryos were vitrified and warmed based on protocols was described previously (Chen et al., 2014). In short, embryo freezing was performed via a Cyrotop carrier system, along with DMSO-EG-S as a cryoprotectant. During thawing, embryos were successively transferred into a dilution solution (1 M–0.5 M–0 M sucrose). Specifically, after surviving the warming procedure, post-thaw day 3 (D3) embryos were cultured at 37^○C in a 6% CO₂, 5% O₂ and 89% N₂ incubator for another 16 h before transfer; this practice was required due to our work schedule. For day 5 (D5) or day 6 (D6) blastocysts, an additional 2–6 h incubation was performed before transfer.

Outcome measures

Outcomes, including perinatal and obstetric complications, were presented as follows: PTB (before 37 completed weeks of gestation), low birth weight (LBW, birth weight <1500 g), macrosomia (birth weight ≥4000 g), gestational diabetes mellitus (GDM, any degree of glucose intolerance with onset or first recognition during pregnancy), premature rupture of membranes (PROM, rupture of the foetal membranes before the onset of labour), postpartum haemorrhage (PPH, greater than 500 ml of vaginal bleeding in 24 h after the delivery of the foetus), placental abruption (premature separation of a normally implanted placenta), placenta previa (placenta completely or partially covering the internal os), placenta accreta (all or partial the placenta are attached directly to the myometrium due to a complete or partial absence of decidua), based on birth weight and the growth standard curves of birth weight, length and head circumference of Chinese newborns of different gestation (Capital Institute of Pediatrics, & Coordinating Study Group of Nine Cities on the Physical Growth and Development of Children, 2020), the babies were categorized into: LBW (birth weight <2500 g), macrosomia (birth weight >4000 g), SGA (birth weight <10th of the national reference), LGA (birth weight >90th of the national reference).

Statistical analysis

We used SPSS software version 26.0 to perform all statistical analysis in this study (SPSS 26.0 software IBM, NY, USA). For the normally distributed continuous variables, the data were given as mean ± standard deviation (SD) and compared using the student t test. For the non-normally distributed continuous data, the data were given as the median (interquartile range) and compared using the Mann-Whitney U test. The chi-square test was used for categorical variables. Adjusted odds ratios (OR) with 95% confidence intervals (CI) were calculated to approximate relative risks of adverse outcomes. Odds ratios, adjusted for maternal age, paternal age, infertility cause, gravidity, parity, pre-pregnancy BMI, number of abortions, infertility duration, type of endometrial preparation, stage of embryo(s) transferred and newborn sex were estimated using multivariate logistic regression. All statistical tests and p values were two-sided, a p value of <0.05 was considered statistically significant.

Results

Participants’ characteristics

A total of 1,609 women who met the inclusion criteria during the study period were included in the analysis. The mean paternal BMI was 24.74 ± 3.55 kg/m², while 43 infants (2.67%) had paternal BMI in the underweight category, 835 infants (51.89%) had paternal BMI in the normal category, 607 infants (37.73%) had paternal BMI in the overweight category, and 124 infants (7.71%) had paternal BMI in the obesity category. The primary parental characteristics and treatment characteristics of the included cycles are illustrated in Table 1. The average age of the males was slightly younger in the underweight group (32.70 ± 4.08) than in the normal weight group (32.89 ± 5.29). The partners of the men among these four groups were in similar age. No significant differences in infertility duration, infertility cause, gravidity, parity, FET endometrial preparation, or stage of embryo transfer were observed among the four paternal BMI categories. As potential confounders, these parameters were included in the logistic regression analysis.

Table 1. Parental and treatment characteristic of frozen embryo transfer (FET) cycles.

Characteristic (n = 1609)	Paternal underweight (n = 43)	Paternal normal weight (n = 835)	Paternal overweight (n = 607)	Paternal obesity (n = 124)	P^a	P^b	P^c
Maternal age (years)^e	32.63 ± 3.46	31.72 ± 3.92	31.50 ± 4.09	31.63 ± 3.77	0.295	0.145	0.802
Maternal BMI (kg/m^2)^e	21.92 ± 3.26	21.77 ± 3.43	21.37 ± 3.25	21.05 ± 2.88	0.154	0.768	0.077
Paternal age (years)^e	32.70 ± 4.08	32.89 ± 5.29	34.30 ± 4.29	33.09 ± 3.88	0.169	0.049	0.523
Paternal BMI (kg/m^2)^e	18.33 ± 1.83	22.37 ± 1.78	27.14 ± 1.65	32.29 ± 2.27	<0.001	<0.001	<0.001
Infertility duration (years)^d	2.0 (3.0–5.0)	2.0 (3.0–4.0)	3.0 (2.0–4.0)	2.0 (3.0–5.0)	0.245	0.476	0.293
Infertility cause
Female	67.44%	69.82%	66.06%	65.32%	0.853	0.741	0.801
	(29/43)	(583/835)	(401/607)	(81/124)
Male	11.63%	12.46%	12.69%	10.48%	0.840	0.873	0.835
	(5/43)	(104/835)	(77/607)	(13/124)
Mixed	20.93%	17.82%	21.25%	24.19%	0.960	0.539	0.663
	(9/43)	(148/835)	(129/607)	(30/124)
Gravidity
1	41.86	51.86%	51.24%	48.39%	0.235	0.201	0.460
	(18/43)	(433/835)	(311/607)	(60/124)
≥2	58.14%	48.14%	48.76%	51.61%
	(25/43)	(402/835)	(296/607)	(64/124)
Parity
0	81.40%	82.75%	84.18%	80.65%	0.630	0.818	0.914
	(35/43)	(691/835)	(511/607)	(100/124)
1	16.28%	15.93%	14.50%	18.55%
	(7/43)	(133/835)	(88/607)	(23/124)
≥2	2.36%	1.32%	1.32%	0.81%
	(1/43)	(11/835)	(8/607)	(1/124)
Number of abortions
0	46.51%	55.09%	55.85%	50.00%	0.234	0.270	0.693
	(20/43)	(460/835)	(339/607)	(62/124)
1	37.21%	25.27%	26.19%	29.84%
	(16/43)	(211/835)	(159/607)	(37/124)
≥2	16.30%	19.64%	17.96%	20.16%
	(7/43)	(164/835)	(109/607)	(25/124)
FET endometrial preparation
NC	20.93%	13.29%	14.99%	16.94%	0.297	0.155	0.557
	(9/43)	(111/835)	(11/607)	(21/124)
OI	6.98%	6.71%	8.07%	5.65%	0.798	0.945	0.751
	(3/43)	(56/835)	(49/607)	(7/124)
AC	72.09%	80%	76.94%	77.42%	0.468	0.209	0.481
	(31/43)	(668/835)	(467/607)	(96/124)
Stage of embryo transferred
Cleavage-stage embryo	25.58%	19.04%	19.93%	24.19%	0.672	0.290	0.178
	(11/43)	(159/835)	(121/607)	(30/124)
Blastocyst	74.42%	80.96%	80.07%	75.81%
	(32/43)	(676/835)	(486/607)	(94/124)
No. of embryos transferred
1	20.93%	18.92%	19.44%	19.35	0.805	0.744	0.909
	(9/43)	(158/835)	(118/607)	(24/124)
2	79.07%	81.08%	80.56%	80.65%
	(34/43)	(677/835)	(489/607)	(100/124)
Embryo quality
Good-quality	49.35%	47.02%	50.91%	47.32	0.058	0.690	0.934
	(38/77)	(711/1512)	(558/1096)	(106/224)
Poor-quality	50.64%	52.97%	49.09%	52.68%
	(39/77)	(801/1512)	(538/1096)	(118/224)

Abbreviations: AC, artificial cycle; BMI, body mass index; NC, natural cycles; OI, ovulation induction.

Data are presented as mean standard deviation for continuous variables and n (%) for dichotomous variables.

^aPaternal overweight vs. normal weight.

^bPaternal underweight vs. normal weight.

^cPaternal obesity vs. normal weight.

^dData are expressed as median (interquartile range) for non-normally distributed continuous variables.

^eData are expressed as mean + SD for normally distributed continuous variables.

Impact of paternal BMI

The incidences of pregnancy and perinatal complications were exhibited in Table 2. The paternal normal group was more likely to suffer from gestational hypertension than the paternal obesity group (2.42% vs. 3.59%), and paternal underweight group was likely to suffer from preeclampsia than other three groups (11.63% vs. 4.43%, 7.57%, 4.03%, respectively). There was no statistically significant difference in the incidence of other obstetric complications among the four groups.

Table 2. Pregnancy, perinatal complications according to paternal BMI.

Items	Paternal underweight (n = 43)	Paternal normal weight (n = 835)	Paternal overweight (n = 607)	Paternal obesity (n = 124)	P^a	P^b	P^c
Pregnancy complications
GDM	32.56%	34.13%	31.96%	29.83%	0.387	0.832	0.345
	(14/43)	(285/835)	(194/607)	(37/124)
Gestational hypertension	6.98%	3.59%	2.47%	2.42%	0.227	0.255	<0.001
	(3/43)	(30/835)	(15/607)	(3/124)
Preeclampsia	11.63%	4.43%	7.57%	4.03%	0.011	0.031	0.840
	(5/43)	(37/835)	(46/607)	(5/124)
Mild preeclampsia	4.65%	2.99%	4.61%	0.81%	0.107	0.539	0.162
	(2/43)	(25/835)	(28/607)	(1/124)
Severe preeclampsia	6.98%	1.44%	2.97	3.23%	0.151	0.006	0.290
	(3/43)	(12/835)	(18/607)	(4/124)
ICP	0%	0.60%	1.48%	0.81%	0.421	0.611	0.555
	(0/43)	(5/835)	(9/607)	(1/124)
Perinatal complications
Placenta previa	6.98%	2.75%	3.95%	4.03%	0.377	0.111	0.436
	(3/43)	(23/835)	(24/607)	(5/124)
Placental abruption	0	0.24%	0.36%	0	0.654	0.748	0.585
		(2/835)	(3/835)
PROM	16.28%	24.31%	20.43%	22.58%	0.082	0.229	0.674
	(7/43)	(203/835)	(124/607)	(28/124)
Abnormal placental cord insertion	4.65%	7.19%	2.64%	4.03%	0.307	0.718	0.808
	(2/43)	(30/835)	(16/607)	(5/124)
Placental adherence	13.95%	12.10%	13.51%	12.90%	0.426	0.716	0.798
	(6/43)	(101/835)	(82/607)	(16/124)
Placenta increta	2.33%	2.28%	1.81%	3.23%	0.543	0.983	0.519
	(1/43)	(19/835)	(11/607)	(4/124)
Postpartum haemorrhage	23.26%	17.13%	15.49%	14.52%	0.407	0.301	0.468
	(10/43)	(143/835)	(94/607)	(18/124)
Polyhydramnios	2.33%	2.99%	3.13%	2.42%	0.882	0.801	0.723
	(1/43)	(25/835)	(19/607)	(3/124)
Oligohydramnios	11.63%	3.11%	4.45%	4.03%	0.184	0.003	0.589
	(5/43)	(26/835)	(27/607)	(5/124)

Abbreviations: ART, Assisted reproductive technology; CI, Confidence interval; GDM, Gestational diabetes mellitus; ICP, Intrahepatic cholestasis of pregnancy; OR, Odds ratio; PROM, premature rupture of membranes; SC, spontaneous conception.

Data are expressed as No. (%).

P^a Paternal overweight vs. normal weight.

P^b Paternal underweight vs. normal weight.

P^c Paternal obesity vs. normal weight.

Table 3 lists the results of neonatal outcomes among the four groups. No significant difference could be found regarding newborn sex, gestational age, LBW, 1 minute Apgar ≤ 7, LGA and SGA. Birthweight among infants in the paternal overweight categories was significantly higher than infants in the paternal normal weight categories. The rate of foetal macrosomia was higher among infants in the paternal overweight (12.36%) category, while lower among infants in the paternal underweight categories (2.33%). There were no differences in the incidence of LBW infants among the four categories. Remarkably, the incidence of LGA infants was significantly higher in the paternal overweight (32.62%) and obesity categories (35.48%) than in the paternal underweight categories (23.26%). The incidence of SGA infants was lower in the paternal overweight categories (1.98%) than in the paternal normal weight categories (3.47%).

Table 3. Neonatal outcomes according to paternal BMI.

Items	Paternal underweight (n = 43)	Paternal normal weight (n = 835)	Paternal overweight (n = 607)	Paternal obesity (n = 124)	P^a	P^b	P^c
Newborn sex
Female (n = 732)	48.84% (21/43)	46.66% (389/835)	43.66% (265/607)	45.96% (57/124)	0.508	0.773	0.745
Male (n = 877)	51.16% (22/43)	53.41% (446/835)	56.34% (342/607)	54.03% (67/124)
Gestational age
<32 W (n = 24)	0	1.32% (11/835)	1.48% (9/607)	3.23% (4/124)	0.741	0.861	0.935
32–36 W (n = 147)	11.63% (5/43)	9.46% (79/835)	8.57% (52/607)	8.87% (11/124)
≥37 W (n = 1438)	88.37% (38/43)	89.22% (745/835)	89.95% (546/607)	87.90% (109/124)
LBW (<2500 g n = 92)	9.30% (4/43)	5.75% (48/835)	4.61% (28/607)	9.68% (12/124)	0.170	0.336	0.943
Macrosomia (≥4000 g n = 161)	2.33% (1/43)	8.38% (70/835)	12.36% (75/607)	12.10% (15/124)	0.013	0.155	0.175
1 min Apgar ≤7	2.33% (1/43)	0.59% (5/835)	0.82% (5/607)	0	0.320	0.180	0.089
5 min Apgar ≤7	2.33% (1/43)	0	0.66% (4/607)	0	0.227	<0.001	0.089
LGA (n = 486)	23.26% (10/43)	28.02% (234/835)	32.62% (198/607)	35.48% (44/124)	0.060	0.496	0.088
SGA (n = 46)	0% (0/43)	3.47% (29/835)	1.98% (12/607)	4.03% (5/124)	0.091	0.214	0.753

Abbreviations: LBW, low birth weight; LGA, large for gestational age; PTB, preterm birth; SGA, small for gestational age.

P^a Paternal overweight vs. normal weight.

P^b Paternal underweight vs. normal weight.

P^c Paternal obesity vs. normal weight.

Impact of paternal BMI stratified by age

As shown in Tables 4 and 5, a stratified analysis by paternal age was conducted. When the man is over 40 years old, paternal normal weight group showed a higher likelihood of gestational hypertension compared to the paternal obesity group (5.88% vs. 0%), the paternal underweight group exhibited a notably higher rate of preeclampsia compared to the other three groups (50% vs. 5.88%, 7.41%, 0%, respectively). No significant difference was observed in the incidence of other obstetric complications among the four groups.

Table 4. Pregnancy, perinatal complications and neonatal outcomes according to paternal BMI with stratified analysis by paternal age.

Items	Paternal underweight (n = 2)	Paternal normal weight (n = 34)	Paternal overweight (n = 27)	Paternal obesity (n = 5)	P^a	P^b	P^c
≥40 year
Pregnancy complications
GDM	0% (0/2)	64.70% (22/34)	55.55% (15/27)	20% (1/5)	0.467	0.068	0.058
Gestational hypertension	0% (0/2)	5.88% (2/34)	0% (0/27)	0% (0/5)	0.200	0.724	0.578
Preeclampsia	50% (1/2)	5.88% (2/34)	7.41% (2/27)	0% (0/5)	0.811	0.095	0.578
Mild preeclampsia	50% (1/2)	0% (0/34)	7.41% (2/27)	0% (0/5)	0.107	0.001	–
Severe preeclampsia	0% (0/2)	5.88% (2/34)	0% (0/27)	0% (0/5)	0.732	0.213	0.589
ICP	0% (0/2)	0% (0/34)	0% (0/27)	0% (0/5)	–	–	–
Perinatal complications
Placenta previa	0% (0/2)	2.94% (1/34)	11.11% (3/27)	0% (0/5)	0.200	0.806	0.698
Placental abruption	0% (0/2)	2.94% (1/34)	0% (0/27)	0% (0/5)	0.369	0.806	0.806
PROM	0% (0/2)	5.88% (2/34)	14.81% (4/27)	20% (1/5)	0.245	0.742	0.269
Abnormal placental cord insertion	0% (0/2)	5.88% (2/34)	7.41% (2/27)	0% (0/5)	0.811	0.724	0.724
Placental adherence	0% (0/2)	20.58% (7/34)	18.51% (5/27)	20% (1/5)	0.840	0.475	0.976
Placenta increta	0% (0/2)	2.94% (1/34)	7.41% (2/27)	20% (1/5)	0.423	0.806	0.106
Postpartum haemorrhage	50% (1/2)	20.58% (7/34)	29.63% (8/27)	40% (2/5)	0.415	0.331	0.336
Polyhydramnios	0% (0/2)	8.82% (3/34)	3.70% (1/27)	0% (0/5)	0.397	0.661	0.489
Oligohydramnios	0% (0/2)	0% (0/34)	0% (0/27)	20% (1/5)	–	–	0.008
Neonatal outcomes
Newborn sex
Female (n = 26)	50% (1/2)	47.06% (16/34)	18.52% (5/27)	80% (4/5)	0.020	0.935	0.169
Male (n = 42) (n = 42)	50% (1/2)	52.94% (18/34)	81.48% (22/27)	20% (1/5)
Gestational age
<32 W (n = 11)	0%	8.82% (3/34)	25.93% (7/27)	20% (1/5)
32–36 W (n = 16)	0%	11.76% (4/34)	37.04% (10/27)	40% (2/5)	0.004	0.475	0.087
≥37 W (n = 41)	100% (2/2)	79.41% (27/34)	37.04% (10/27)	40% (2/5)
Low birthweight (<2500 g) (n = 92)	0%	2.94% (1/34)	3.70% (1/27)	0%	0.868	0.806	0.698
Macrosomia (≥4000 g) (n = 161)	0%	5.88% (2/34)	7.41% (2/27)	20% (1/5)
1 min Apgar ≤ 7	–	–	–	–	–	–	–
5 min Apgar ≤ 7	–	–	–	–	–	–	–
LGA (n = 27)	50% (1/2)	23.53% (8/34)	41.18% (14/34)	80% (4/5)	0.120	0.401	0.011
SGA (n = 2)	0%	5.88% (2/34)	0%	0%	0.200	0.724	0.578

Abbreviations: ART, Assisted reproductive technology; CI, Confidence interval; GDM, Gestational diabetes mellitus; ICP, Intrahepatic cholestasis of pregnancy; OR, Odds ratio; LBW, low birth weight; LGA, large for gestational age; PROM, premature rupture of membranes; PTB, preterm birth; SC, spontaneous conception; SGA, small for gestational age.

Data are expressed as No. (%).

P^a Paternal overweight vs. normal weight.

P^b Paternal underweight vs. normal weight.

P^c Paternal obesity vs. normal weight.

Table 5. Pregnancy, perinatal complications and neonatal outcomes according to paternal BMI with stratified analysis by paternal age.

Items	Paternal underweight (n = 2)	Paternal normal weight (n = 34)	Paternal overweight (n = 27)	Paternal obesity (n = 5)	P^a	P^b	P^c
<40 years
Pregnancy complications
GDM	34.15% (14/41)	32.83% (263/801)	30.86% (179/580)	30.25% (36/119)	0.438	0.862	0.315
Gestational hypertension	7.32% (3/41)	3.49% (28/801)	2.59% (15/580)	2.52% (3/119)	0.337	0.205	0.582
Preeclampsia	9.76% (4/41)	4.37% (35/801)	7.59% (44/580)	4.20 (5/119)	0.011	0.109	0.933
Mild preeclampsia	2.44% (1/41)	3.12% (25/801)	4.48% (26/580)	0.84% (1/119)	0.185	0.805	0.161
Severe preeclampsia	7.32% (3/41)	1.25% (10/801)	3.10% (18/580)	3.36% (4/119)	0.016	0.002	0.079
ICP	0% (0/41)	0.62% (5/801)	1.55% (9/580)	0.84% (1/119)	0.089	0.612	0.785
Perinatal complications
Placenta previa	7.31% (3/41)	2.75% (22/801)	3.62% (21/580)	4.20% (5/119)	0.356	0.093	0.380
Placental abruption	0% (0/41)	0.12% (1/801)	0.52% (3/580)	0% (0/119)	0.180	0.821	0.700
PROM	17.07% (7/41)	25.09% (201/801)	20.69% (120/580)	22.69% (27/119)	0.056	0.245	0.571
Abnormal placental cord insertion	4.88% (2/41)	3.49% (28/801)	2.41% (14/580)	4.20% (5/119)	0.248	0.641	0.699
Placental adherence	14.63% (6/41)	11.74% (94/801)	13.28% (77/580)	12.61% (15/119)	0.391	0.576	0.784
Placenta increta	2.44% (1/41)	2.25% (8/801)	1.55% (9/580)	2.52% (3/119)	0.357	0.936	0.852
Postpartum haemorrhage	21.95% (9/41)	16.98% (136/801)	14.83% (86/580)	13.45% (16/119)	0.283	0.411	0.333
Polyhydramnios	2.44% (1/41)	2.75% (22/801)	3.10% (18/580)	2.52% (3/119)	0.696	0.906	0.888
Oligohydramnios	12.19% (5/41)	3.25% (26/801)	4.66% (27/580)	3.36% (4/119)	0.178	0.003	0.947
Neonatal outcomes
Newborn sex
Female (n = 706) (n = 706)	48.78% (20/41)	46.57% (373/801)	44.83% (260/580)	44.54% (53/119)	0.522	0.782	0.679
Male (n = 835) (n = 835)	51.22% (21/41)	53.43% (428/801)	55.17% (320/580)	55.46% (66/119)
Gestational age
<32 W (n = 13)	0%	0.99% (8/801)	0.34% (2/580)	2.52% (3/119)
32–36 W (n = 131)	12.19% (5/41)	9.36% (75/801)	7.24% (42/580)	7.56% (9/119)	0.078	0.708	0.926
≥37 W (n = 1397)	87.80% (36/41)	89.63% (718/801)	92.41% (536/580)	86.29% (107/119)
Low birthweight (<2500 g) (n = 90)	9.75% (4/41)	5.86% (47/801)	4.66% (27/580)	10.08% (12/119)	0.323	0.309	0.080
Macrosomia (≥4000 g) (n = 156)	2.44% (1/41)	8.49% (68/801)	12.59% (73/580)	11.76% (14/119)	0.013	0.168	0.242
1 min Apgar ≤ 7	2.44% (1/41)	0.62% (5/801)	0.86% (5/580)	0%	0.411	0.178	0.791
5 min Apgar ≤ 7	2.44% (1/41)	0%	0.68% (4/580)	0%	0.019	<0.001	–
LGA (n = 459)	21.95% (9/41)	28.21% (226/801)	31.72% (184/580)	33.61% (40/119)	0.159	0.383	0.225
SGA (n = 44)	0%	3.37% (27/801)	2.07% (12/580)	4.20% (5/119)	0.149	0.232	0.644

Abbreviations: ART, Assisted reproductive technology; CI, Confidence interval; GDM, Gestational diabetes mellitus; ICP, Intrahepatic cholestasis of pregnancy; OR, Odds ratio; LBW, low birth weight; LGA, large for gestational age; PROM, premature rupture of membranes; PTB, preterm birth; SC, spontaneous conception; SGA, small for gestational age.

Data are expressed as No. (%).

P^a Paternal overweight vs. normal weight.

P^b Paternal underweight vs. normal weight.

P^c Paternal obesity vs. normal weight.

In neonatal outcomes, the incidence of LGA infants was significantly higher in the paternal overweight (41.48%) and obesity categories (80%) than in the paternal normal categories (23.53%). When the man is under 40 years old, the incidence of GDM was similar across all four paternal groups; a higher incidence of preeclampsia was observed in the paternal underweight group (9.76%) compared to the normal weight (4.37%), overweight (7.59%), and obesity groups (4.20%). A higher incidence of preeclampsia was observed in the paternal underweight group (9.76%) compared to the normal weight (4.37%), overweight (7.59%), and obesity groups (4.20%). No significant difference could be found regarding newborn sex, gestational age, LBW, 1-minute Apgar ≤ 7, LGA, and SGA. The rate of foetal macrosomia was higher among infants in the paternal overweight (12.59%) category, while lower among infants in the paternal underweight categories (2.44%).

Impact of pre-pregnancy paternal BMI

The results of the logistic regression analysis are summarized in Tables 6 and 7. After adjustments were made for confounders, an increased parental BMI was associated with preeclampsia, the incidence of pre-eclampsia in the paternal overweight categories were significantly higher compared with those among paternal normal weight (aOR 1.830, 95% CI 1.165–2.874), paternal underweight categories also exhibited higher rates of preeclampsia (aOR 2.970, 95% CI 1.099–8.029). Differences in the other maternal parameters, including the incidences of GDM, ICP, PROM and placental adherence, were not observed in any of the four categories. The incidence of macrosomia in the paternal overweight categories (aOR 1.527, 95% CI 1.078–2.163) were significantly higher than those among normal controls after adjustment for known confounding factors. The rates of LGA babies were higher in the paternal overweight category (aOR 1.260, 95% CI 1.001–1.587), compared with those in the paternal normal weight before and after adjusting. No association was seen between paternal pre-pregnancy BMI and the risk of other neonatal outcomes.

Table 6. Associations between parental pre-pregnancy BMI and maternal outcomes of singletons in multilevel logistic regression analyses.

Items	Paternal underweight^a(n = 43)	P	Paternal overweight^a (n = 607)	P	Paternal obesity^a (n = 124)	P
GDM
Crude OR (95% CI)	0.932 (0.485–1.791)	0.832	0.907 (0.726–1.133)	0.388	0.821 (0.544–1.237)	0.345
AOR (95% CI)	0.894 (0.442–1.808)	0.754	0.888 (0.616–1.281)	0.525	0.748 (0.378–1.479)	0.404
Gestational hypertension	6.97%		2.47%		2.42%
Crude OR (95% CI)	2.012 (0.589–6.875)	0.265	0.680 (0.363–1.275)	0.229	0.665 (0.200–2.213)	0.506
AOR(95% CI)	1.962 (0.573–6.722)	0.283	0.667 (0.355–1.252)	0.207	0.654 (0.196–2.184)	0.490
Preeclampsia
Crude OR (95% CI)	2.838 (1.056–7.629)	0.039	1.768 (1.132–2.763)	0.012	0.906 (0.349–2.351)	0.840
AOR(95% CI)	2.970 (1.099–8.029)	0.032	1.830 (1.165–2.874)	0.009	0.978 (0.375–2.549)	0.963
ICP
Crude OR (95% CI)	–	–	2.498 (0.833–7.492)	0.102	1.350 (0.156–11.648)	0.785
AOR(95% CI)	–	–	2.393 (0.795–7.202)	0.121	1.343 (0.155–11.650)	0.789
Placenta previa
Crude OR (95% CI)	2.648 (0.763–9.189)	0.125	1.453 (0.812–2.600)	0.208	1.483 (0.553–3.976)	0.433
AOR(95% CI)	2.635 (0.754–9.211)	0.129	1.510 (0.836–2.726)	0.172	1.576 (0.582–4.265)	0.371
Placental abruption
Crude OR (95% CI)	–	–	2.069 (0.345–12.418)	0.427	–	–
AOR(95% CI)	–	–	3.863 (0.399–37.413)	0.243	–	–
PROM
Crude OR (95% CI)	0.605 (0.265–1.381)	0.233	0.799 (0.621–1.029)	0.082	0.908 (0.579–1.424)	0.674
AOR(95% CI)	0.629 (0.275–1.440)	0.273	0.786 (0.608–1.014)	0.064	0.945 (0.600–1.489)	0.807
Abnormal placental cord insertion
Crude OR (95% CI)	1.309 (0.302–5.666)	0.719	0.726 (0.392–1.345)	0.309	1.127 (0.429–2.963)	0.808
AOR(95% CI)	1.240 (0.284–5.406)	0.775	0.696 (0.375–1.291)	0.250	1.139 (0.431–3.006)	0.793
Placental adherence
Crude OR (95% CI)	1.178 (0.485–2.862)	0.717	1.135 (0.831–1.551)	0.426	1.077 (0.612–1.894)	0.798
AOR(95% CI)	1.169 (0.475–2.874)	0.734	1.215 (0.882–1.678)	0.234	1.104 (0.620–1.968)	0.735
Placenta increta
Crude OR (95% CI)	1.023 (0.134–7.822)	0.983	0.293 (0.374–1.678)	0.544	1.432 (0.479–4.280)	0.521
AOR(95% CI)	1.059 (0.137–8.186)	0.957	0.812 (0.379–1.741)	0.592	1.400 (0.459–4.272)	0.554
Postpartum haemorrhage
Crude OR (95% CI)	1.466 (0.707–3.043)	0.304	0.887 (0.667–1.178)	0.407	0.822 (0.483–1.398)	0.469
AOR(95% CI)	1.493 (0.716–3.113)	0.285	0.912 (0.684–1.216)	0.531	0.780 (0.451–1.348)	0.373
Polyhydramnios
Crude OR (95% CI)	0.771 (0.102–5.831)	0.801	1.047 (0.571–1.919)	0.882	0.803 (0.239–2.701)	0.723
AOR(95% CI)	0.731 (0.096–5.554)	0.762	1.029 (0.559–1.893)	0.928	0.751 (0.221–2.546)	0.645
Oligohydramnios
Crude OR (95% CI)	4.094 (1.490–11.251)	0.006	1.448 (0.837–2.508)	0.186	1.307 (0.496–3.470)	0.591
AOR(95% CI)	4.432 (1.594–12.328)	0.004	1.396 (0.804–2.423)	0.236	1.311 (0.492–3.494)	0.588

Abbreviations: CI, Confidence interval; GDM, Gestational diabetes mellitus; ICP, Intrahepatic cholestasis of pregnancy; OR, Odds ratio; PROM, premature rupture of membranes.

^aPaternal normal weight (n = 835) was used as the reference.

Table 7. Associations between parental pre-pregnancy BMI and neonatal outcomes of singletons in multilevel logistic regression analyses.

Items	Paternal underweight^a (n = 43)	P	Paternal overweight^a (n = 607)	P	Paternal obesity^a (n = 124)	P
PTB
Crude OR (95% CI)	1.131 (0.434–2.950)	0.801	0.926 (0.653–1.312)	0.644	1.094 (0.601–1.992)	0.768
AOR(95% CI)	1.165 (0.445–3.046)	0.756	0.926 (0.491–1.279)	0.669	1.119 (0.612–1.316)	0.715
LBW
Crude OR (95% CI)	1.682 (0.577–4.900)	0.341	0.793 (0.491–1.273)	0.342	1.757 (0.905–3.408)	0.096
AOR(95% CI)	1.674 (0.573–4.891)	0.347	0.788 (0.488–1.273)	0.330	1.822 (0.935–3.550)	0.078
Macrosomia
Crude OR (95% CI)	0.260 (0.035–1.919)	0.187	1.541 (1.092–2.173)	0.014	1.504 (0.831–2.720)	0.177
AOR(95% CI)	0.272 (0.037–2.007)	0.202	1.527 (1.078–2.163)	0.017	1.514 (0.832–2.753)	0.174
LGA
Crude OR (95% CI)	0.778 (0.378–1.604)	0.497	1.243 (0.991–1.560)	0.060	1.413 (0.949–2.102)	0.089
AOR(95% CI)	0.779 (0.377–1.611)	0.501	1.260 (1.001–1.587)	0.049	1.458 (0.974–2.183)	0.067
SGA
Crude OR (95% CI)	–	–	0.561 (0.284–1.108)	0.096	1.168 (0.443–3.076)	0.754
AOR(95% CI)	–	–	0.557 (0.281–1.103)	0.093	1.214 (0.459–3.212)	0.696

Abbreviations: PTB, preterm birth; LBW, low birth weight; LGA, large for gestational age; SGA, small for gestational age.

^aPaternal normal weight (n = 835) was used as the reference.

Discussion

In our study, we found that paternal BMI has an independent impact on the birthweight of singletons after FET cycles. Paternal overweight and obesity may be associated with macrosomia and LGA, and the paternal underweight group were likely to suffer from pre-eclampsia than other three groups.

The effect of male BMI on ART outcomes is less studied, but earlier results do suggest, in line with this study, a negative impact of increasing male BMI on ART outcomes. A study showed that both overweight and obese men were more likely to experience infertility compared with normal weight men (Campbell & McPherson, 2019). In a retrospective analysis of 305 couples undergoing ART, increased male BMI was associated with significantly reduced live birth rates (Bakos et al., 2011). In natural conception cycles, though two studies reported on diametrically opposed birthweight outcomes (macrosomia (Yang et al., 2015) and SGA (McCowan et al., 2011)) although, interestingly, both of them reported significant effects. Nonetheless, the included studies on the association between paternal BMI and the development of offspring weight or BMI suggested that when fathers with higher BMI, their children could develop higher BMIs. Specifically, both paternal overweight and obesity emerged as independent predictors of giving birth to LGA infants post-FET. Paternal overweight alone was identified as an isolated risk factor for producing infants with macrosomia or very LGA after these cycles (Ma et al., 2020). Although it combined maternal and paternal BMI as one risk factor, Wang et al. (2016) observed significant differences in birthweight and the rates of foetal macrosomia of singletons among the different BMI categories after fresh ET cycles.

The finding of increased infertility diagnoses of men with increasing BMI mirrors those findings of the previous systematic review assessing male BMI and sperm function (Campbell et al., 2015). This is likely due to obesity’s negative effects on sex hormones (the hypothalamic–pituitary–gonadal (HPG) axis) and sperm function. Men with an increasing BMI are more likely to have reduced plasma concentrations of testosterone and increased concentrations of oestrogen, both are independently associated with subfertility and reduced sperm counts via disrupting spermatogenesis (Cohen, 1999; Jensen et al., 2004). In addition, the findings that increased male BMI could increase rates of both SGA and macrocosmia mirror those for animal models. A rat model of male obesity and metabolic syndrome found that male rats fed a diet high in fat produced smaller offspring at birth (Ng et al., 2010), while a mouse model of male obesity prior to the onset of metabolic syndrome found that male mice fed a high fat diet produced larger offspring on post-natal day 3 (Fullston et al., 2013).

In an analysis of 301 IVF cases exploring the relationship between male BMI and ART outcomes, the study found that elevated male BMI was associated with a negative impact on pregnancy rates, potentially mediated by compromised embryo quality (Anifandis et al., 2013). Another study demonstrated the impact of paternal overweight on male reproductive health, as these patients had a higher percentage of immature sperm with impaired chromatin integrity in their semen and had a decreased fertilization rate, cumulative live birth rate following assisted reproductive treatments (Bibi et al., 2022). Research indicates that paternal obesity alters the expression of sperm-specific microRNAs while also promoting histone modification and DNA methylation in germ cells (Zhao et al., 2017). In spermatozoa, paternal obesity has been shown to elevate histone H3 occupancy in gene promoters linked to embryogenesis, as well as enhance monomethylation at lysine 4 on histone H3 (H3K4me1) in genes that regulate embryonic development (Terashima et al., 2015). In populations with pre-conception paternal obesity, epigenetic alterations in leptin genes have been observed in offspring (Masuyama et al., 2016). Leptin serves as a trophic factor in the formation of hypothalamic circuits, which govern energy balance as well as food-seeking and reward behaviours (Jiménez-Chillarón et al., 2012). Collectively, these alterations may contribute to abnormal birth weight and other metabolic disorders. Additionally, insulin-like growth factor-II (IGF-II), which regulates both placental and foetal growth, is imprinted and expressed from the paternal copy of the gene. However, findings from animal-based experiments should not be directly applied to human couples seeking assisted reproductive treatments. Further investigations involving larger human cohorts are necessary to clarify these issues in the context of assisted reproduction.

Although the underlying mechanisms for the impact of paternal BMI on neonatal outcomes remain unknown, it is quite likely that epigenetic changes in spermatozoa induce paternal programming of the neonatal phenotype. McPherson et al. (2014) suggest that paternal overweight/obesity induces paternal programming of offspring phenotypes, which is likely mediated through genetic and epigenetic changes in spermatozoa. During spermatogenesis, environmental exposures such as diet, lifestyle, and other exposures can lead to irreversible epigenetic changes and phenotypic consequences expressed in the following generation (N. Li et al., 2016).

There are some limitations to in our study. First, the database did not record environmental and occupational information, as well as, unhealthy lifestyle and psychological status of patients, which may contribute to the miscarriage. Second, we did not investigate sperm parameters or any significant sperm parameters (such as Sperm DNA Fragmentation or oxidative stress) so as to see whether paternal BMI has any effect through the sperm parameters, one study indicated that semen parameters are more strongly correlated with male age than with BMI, implicating age as a more influential factor in IVF outcomes (Paasch et al., 2010). However, other research contends that the impact of age on semen parameters is not consistently evident (Jensen et al., 2004). Given that all female participants in this study had BMIs within the normal range, the relationship between couples’ BMI and ART outcomes remains underexplored. Future studies are warranted to address this gap.

In summary, our study indicates that parental pre-pregnancy overweight or obesity has an adverse effect on the perinatal complications and neonatal outcomes. Specifically, increased parental pre-pregnancy BMI, increases the odds of LGA. Considering that abnormal birthweight elevates the risk of complications during the foetal and neonatal periods and many diseases in adulthood, it is premature to say whether the differences in birthweight that we discovered in this study have any clinical significance. These findings should be confirmed by future large prospective studies.

Statement of Ethics

The study was approved by the local Ethics Committee of our hospital and was performed after obtaining informed consent from each patient.

Funding sources

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by National Natural Science Foundation of China (81871210) and Natural Science Foundation of Jiangsu Province (BK20171126).

Acknowledgments

The authors acknowledge the physicians, nurses, and scientific staff of Department of Reproductive Medicine, Nanjing Maternity and Child Health Care Hospital.

Disclosure statement

No potential conflict of interest was reported by the authors.