- Email: [email protected]
Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis
ARTICLE IN PRESS
Review
Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis Rabia Mushtaq a,*, Jyotsna Pundir b, Chiara Achilli c, Osama Naji a, Yacoub Khalaf a, Tarek El-Toukhy a a Assisted Conception Unit, Guy’s and St Thomas Hospital NHS Foundation Trust, 11th Floor, Tower Wing, London, UK b Centre of Reproductive Medicine, St Bartholomew’s Hospital, London, UK c The Hewitt Fertility Centre, Liverpool Women’s Hospital, Liverpool, UK
Ms Rabia Mushtaq (MBBS, FCPS, MRCOG) currently works at Milton Keynes University Hospital, and developed an interest in research while working as Clinical Research Fellow at Guy’s and St Thomas Foundation Trust in assisted conception. Her special interests are factors influencing IVF and pregnancy outcomes. KEY MESSAGE Published research suggests that raised body mass index (BMI) has a negative effect on IVF and intracytoplasmic sperm injection treatment outcome. Future well-designed, robust prospective studies adhering to the World Health Organization definitions of BMI categories and considering important confounding variables are needed to confirm our study results.
A B S T R A C T Men with a body mass index (BMI) of 30 or over are more likely to have reduced fertility and fecundity rates. This systematic review and meta-analysis evaluated the effect of male BMI on IVF and intracytoplasmic sperm injection (ICSI) outcome. An electronic search for published literature was conducted in MEDLINE and EMBASE between 1966 and November 2016. Outcome measures were clinical pregnancy rates (CPR) and live birth rates (LBR) per IVF or ICSI cycle. Eleven studies were identified, including 14,372 cycles; nine reported CPR and seven reported LBR. Pooling of data from those studies revealed that raised male BMI was associated with a significant reduction in CPR (OR 0.78, 95% CI 0.63 to 0.98, P = 0.03) and LBR (OR 0.88, 95% CI 0.82 to 0.95, P = 0.001) per IVF–ICSI treatment cycle. Male BMI could be an important factor influencing IVF–ICSI outcome. More robust studies are needed to confirm this conclusion using standardized methods for measuring male BMI, adhering to the World Health Organization definitions of BMI categories, accounting for female BMI, IVF and ICSI cycle characteristics, including the number of embryos transferred and embryo quality, and use the live birth rate per cycle as primary outcome. © 2018 Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. E-mail address: [email protected] (R Mushtaq). https://doi.org/10.1016/j.rbmo.2018.01.002 1472-6483/© 2018 Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
2
Introduction Obesity is a global pandemic, which carries major health consequences and negative effect on quality of life. Worldwide obesity has more than doubled since 1980 (Nguyen and El-Serag, 2010), and is projected to increase further owing to anticipated demographic changes (Popkin et al, 2012). Male factors are thought to contribute to up to 50% of subfertility cases, with 31.5% being attributed solely to a male factor cause (Agarwal et al., 2015; Thonneau et al., 1991). Men who have a body mass index (BMI, measured as kg/m2) of 30 or more are more likely to have reduced fertility and fecundity rates (NICE, 2013; Sundaram et al., 2017). Several mechanisms have been attributed to the reduced fecundity in obese men, including lower serum testosterone and raised serum oestradiol levels, impaired spermatogenesis and erectile dysfunction (Andersson et al., 2008; Katib, 2016). Recent data evaluating the relationship between waist circumference and semen parameters in men without known infertility, indicated higher prevalence of lower ejaculatory volume and oligozoospermia in men with increasing waist circumference and BMI (Eisenberg et al., 2014). In addition, lower sperm concentration and higher percentage of abnormal sperm morphology were associated with increasing body adiposity in a recent cross-sectional cohort study (Tsao et al., 2015). Evidence also shows that raised male BMI could be associated with a lower success rate after IVF treatment. A large cohort study showed that couples in whom both partners were either overweight or obese had the lowest odds for live birth after IVF (Petersen et al., 2013). Furthermore, a large observational study demonstrated that male partner BMI had a greater effect on embryo quality and IVF outcome than semen analysis parameters (Anifandis et al., 2013). Increased level of sperm DNA damage associated with male obesity has been shown to be related to lower pregnancy and higher miscarriage rates in both IVF and ICSI cycles (Zhao et al., 2014). The results of published studies addressing the relationship between raised male BMI and IVF–ICSI outcome, however, are conflicting, with some studies reporting a negative effect of raised male BMI (Bakos et al., 2011; Merhi et al., 2013; Umul et al., 2015), and others reporting no effect (Braga et al., 2012; Schliep et al., 2015; Thomsen et al., 2014). In this study, we sought to systematically review and summarize the existing evidence related to the effect of male BMI on clinical pregnancy and live birth rates after IVF–ICSI treatment.
Materials and methods Literature search methodology Electronic searches for published literature in MEDLINE, and EMBASE were conducted from database inception until November 2016 to capture citations including male partner BMI and reproductive outcome after IVF–ICSI treatment. A combination of medical subject headings (MeSH) and text words were used to generate two subsets of citations: one including ‘male body mass index’ (paternal obesity, BMI, paternal body mass index, male adiposity, overweight men) and the second subset for assisted reproductive techniques (assisted reproductive technology, assisted reproduction, ART cycles, IVF, ICSI, in
vitro fertilisation, invitro fertilisation, in-vitro fertilization, invitro fertilization and intra-cytoplasmic sperm injection). These subsets were combined with ‘AND’ to generate a subset of citations relevant to our research question. No language restrictions were applied. The reference lists of all known primary and review articles were examined to identify relevant articles not captured by electronic searches.
Study selection and outcome measures We considered prospective and retrospective cohort studies, published in full or as abstracts, examining the effect of male BMI on the outcome of IVF and ICSI treatment. Participants were male partners of couples undergoing IVF or ICSI, for whom information about male BMI was available and who were using ejaculated sperm for IVF or ICSI treatment. Studies involving natural conceptions, oocyte donation cycles, intrauterine insemination, ovulation induction and those not using the World Health Organization (WHO) criteria for BMI reporting were excluded. Studies were selected in a two-stage process. In the first instance, two reviewers (RM and CA) independently scrutinized all the titles and abstracts from the electronic searches and full manuscripts of all citations that definitely, or possibly, met the predefined selection criteria were retrieved. After examining the full manuscripts, final inclusion or exclusion decisions were made. Any disagreement about inclusion was resolved by consensus after consultation with a third reviewer (JP). The outcome measures considered for this review were clinical pregnancy and live birth rates per cycle started.
Data extraction and quality assessment Data extraction and quality assessment was carried out independently by two reviewers (RM and CA). For each study included, information was obtained on population size, study design, male BMI categories used, number of participants in each category and study outcome measures. Duplicate studies or studies with overlapping populations were excluded. In cases of missing or unclear data, the authors of the primary studies were contacted. The selected studies were assessed for methodological quality by using the components of study design that are related to internal validity (Centre for Reviews and Dissemination, 2001). Meta-analysis of observational studies in epidemiology (MOOSE) guidelines were followed (Stroup et al., 2000). The methodological quality and risk of bias of each study was assessed using the Newcastle–Ottawa quality assessment scale (NOS) (Wells et al., 2000). A study with NOS score of 6 or higher was regarded as a high-quality study. We combined data from studies if they had similar design, intervention and outcome measures. Data collected compared the clinical pregnancy rate (CPR) and live birth rate (LBR) in the study group (male BMI of 25 or more, i.e. overweight and obese men) with a control group of men with a normal BMI (18.5–24.9). A sensitivity analysis on overweight (BMI 25–29.9), obese (BMI 30–34.9) and morbidly obese (BMI ≥35) men was carried out, and each subgroup was compared with the normal BMI group (BMI 18.5–24.9) for both outcome measures. Similarly, studies that separated data for IVF and ICSI cycles were analysed in another subgroup analysis for the CPR and LBR. A further analysis was carried out including only studies that accounted for female BMI.
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
Statistical analysis From each study, binary data were extracted in 2 × 2 tables, and the results were pooled and summarized as relative risks (RR) with 95% confidence interval (CI) using fixed-effects (Mantel and Haenszel, 1959) or random-effects model as appropriate (DerSimonian and Laird, 1986). Heterogeneity of the exposure effects was evaluated graphically using forest plots (Lewis and Clarke, 2001) and statistically using the I2 statistic to quantify heterogeneity across studies (Higgins and Thompson, 2002). Substantial heterogeneity was considered if the I2 was more than 50% and considered when interpreting the results. Clinical heterogeneity was analysed based on the variation in features of the population, intervention and study quality. RevMan 5.3 software (The Cochrane Collaboration, Oxford, UK) was used for all statistical analyses. To assess for publication bias, a funnel plot analysis using the Egger test was conducted (Egger et al., 1997).
Results Study identification and selection is summarized in Figure 1. The search strategy yielded 92 citations captured from the electronic search, and 20 were obtained from review of reference list of original and review articles. Of these 112 studies, 19 duplicates were removed. Of the remaining 93 studies, 78 were excluded as it was clear from the title or abstract that they did not fulfil the selection criteria. Full manuscripts were retrieved for the remaining 15 studies for detailed evaluation. Four articles (Braga et al., 2012; Kupka et al., 2010; Link et al., 2012; Wu et al., 2015) were then excluded (Table 1) and
Figure 1 – Study identification and selection (11 articles included in systematic review).
3
Table 1 – Excluded studies. Study
Reason
Kupka et al. (2010)
The comparison was only between BMI under 30 and over 30. As both normal weight and overweight men were categorized together, data could not be used. Linear data, interpreted as increasing BMI, rather than WHO BMI categories (WHO, 2017). Definition of normal weight, overweight and obese was not in accordance with WHO criteria. Abstract only, more robust data inclusive of the period covered by this study was included by Schliep et al. (2015). This study is not included in the analysis, because of the risk of duplicating data.
Braga et al. (2012) Wu et al. (2015) Link et al. (2012)
BMI, body mass index; WHO, World Health Organization.
11 studies (Anifandis et al., 2013; Bakos et al., 2011; Colaci et al., 2012; Keltz et al., 2010; Merhi et al., 2013; Mushtaq et al., 2016; Petersen et al., 2013; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015; Wang et al., 2016) were included in this systematic review and meta-analysis (Table 2).
Study characteristics The 11 studies selected in our review included 14,372 IVF–ICSI cycles. Of those 11 studies, seven reported both CPR and LBR (Anifandis et al., 2013; Bakos et al., 2011; Colaci et al., 2012; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015; Wang et al., 2016). Three studies, however, reported only the CPR (Keltz et al., 2010, Merhi et al., 2013, and Mushtaq et al., 2016) and one study reported only the LBR (Petersen et al., 2013). The characteristics of the 11 selected trials are presented in Table 2. Keltz et al. (2010) retrospectively assessed the CPR in 290 IVF– ICSI cycles, and concluded that raised male BMI adversely affected the CPR after IVF, but not ICSI, cycles. Bakos et al. (2011) conducted a retrospective analysis of 305 fresh IVF–ICSI cycles, and reported that raised male BMI was associated with poorer blastocyst development and reduced CPR and LBR. Colaci et al. (2012) prospectively studied 172 cycles in 114 couples and concluded that raised male BMI had no effect on IVF–ICSI outcome. The study stated outcomes in IVF and ICSI categories separately as well as in BMI categories, and expressed it as odds ratio rather than numbers for each category. Therefore, data from this study was included only in subgroup analysis and not included in main groups, i.e, CPR and LBR in total. The retrospective studies of Anifandis et al. (2013) and Merhi et al. (2013) included 301 and 344 cycles respectively, and both concluded that raised male BMI negatively influenced pregnancy rates after IVF–ICSI cycles, although they reported different effects on embryo quality. Petersen et al. (2013) reported a population-based cohort study of 25,191 IVF–ICSI cycles and concluded the raised male BMI negatively influenced IVF and, to a lesser extent, ICSI outcome. Petersen et al. (2013) and Anifandis et al. (2013) presented data in four group categories. Only groups with female BMI less than 25 were included in the analysis. Similarly, Umul et al. (2015) retrospectively analysed 177 ICSI cycles and reported that raised male BMI was associated with lower implantation, pregnancy and live birth rates. In contrast, Thomsen et al. (2014) and Schliep et al. (2015) prospectively studied 612 and 735 fresh IVF–ICSI cycles, respectively, and concluded that raised male BMI had no effect on IVF–ICSI outcome.
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
4
Study type and number of participants/cycles
BMI group and numbers included in analysis
Inclusion criteria
Exclusion criteria
Outcome measures
Comments
Keltz et al. (2010)
Retrospective Cycles (n = 290); IVF–ICSI.
Fresh cycles.
Donor Frozen cycles.
Semen parameters; day-3 embryos; numbers; implantation rate; CPR.
CPR was defined as intrauterine gestational sac on transvaginal ultrasound; number of couples not mentioned.
Bakos et al. (2011)
Retrospective men (n = 305); IVF–ICSI; no difference in maternal BMI for male grouping of normal, overweight, or obese. In morbidly obese male BMI group, maternal BMI was higher compared with normal-weight group (P < 0.01). Prospective cycles (n = 172) Couples undergoing IVF–ICSI (n = 114).
<25 (n = 62 cycles); ≥ 25 (n = 228 cycles) Total (n = 290 cycles); data were adjusted for female BMI in results. Men/cycles < 25 (n = 63); 25–30 (n = 148); 30–35 (n = 62); > 35 (n = 32); total (n = 305).
Female partner aged <38 years; fresh cycle included once only in study; therefore, number of men is same as number of cycles.
Donor or frozen sperm treatment; andrological dysfunction.
Semen parameters; fertilization rate; embryo quality; implantation rate; positive HCG; CPR; LBR; miscarriage rate.
GnRH antagonist protocol. Clinical pregnancy indicated by the presence of a viable fetal heart beat on ultrasound 4–6 weeks after embryo transfer.
More than one cycle/couple included.
Donor.
Semen parameters; fertilization rate; embryo quality ; implantation rate; CPR; LBR.
Fresh cycle included once only in study; therefore, number of men is same as number of cycles.
Not mentioned.
Semen parameters; embryo quality; implantation rate; positive HCG; CPR; LBR.
Zygote pronuclear score; day-3 embryo numbers; day-3 grade for transferred embryos; CPR. Live birth.
CPR was defined as presence of intrauterine pregnancy confirmed by ultrasound; LBR was defined as birth of a neonate after 24 weeks of gestation. Clinical pregnancy indicated by intrauterine sac seen by ultrasound 3–4 weeks after HCG; subgroup analysis for overweight and obese was not included, neither was analysis for IVF or ICSI; day of embryo transfer not documented; BMI of female partner was controlled for. All sperm injected into oocytes were motile.
Colaci et al. (2012)
<25 (n = 54 cycles); 25–30 (n = 80) cycles > 30 (n = 38) cycles Total (n = 172); data were adjusted for female BMI in results. Men BMI >25 and women normal BMI, (n = 142 couples); both men and women with normal BMI (n = 64); total (n = 206).
Anifandis et al. (2013)
Total retrospective men undergoing IVF–ICSI (n = 301).
Merhi et al. (2013)
Retrospective cycles of IVF–ICSI (n = 344)
<25 (n = 75 cycles); ≥25 (n = 269 cycles); total (n = 344 cycles); data were adjusted for female BMI in results.
Fresh cycles.
Donor Frozen cycles.
Petersen et al. (2013)
Population-based cohort study; couples (n = 12566); IVF–ICSI cycles for which male BMI information was available (n = 1906).
Men and women < 18.5 (n = 9); cycles (not included in Figure 7 analysis); 18.5–25 (n = 905 cycles); > 25–30 (n = 742 cycles); > 30 (n = 250 cycles); total (n = 1906 Only cycles where female BMI was normal were included in analysis in Figures 3,5,9 and 11.
Autologous oocytes.
Donor.
Increased BMI in either male or female negatively influence live birth rate, and this association is less clear with ICSI.
(continued on next page)
ARTICLE IN PRESS
Studies
REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
Table 2 – Characteristics of included studies.
Study type and number of participants/cycles
BMI group and numbers included in analysis
Inclusion criteria
Exclusion criteria
Outcome measures
Comments
Thomsen et al. (2014)
Prospective; (n = 612 couples in total); 86 split cycle (IVF + ICSI); 167 IUI were excluded from analysis; IVF, 233 started IVF and 126 ICSI. Couples in final analysis who had oocyte retrieval 228 + 124 ICSI =352 in total.
< 25 (n = 148 men); 25–30 (n = 160 men); > 30 (n = 44 men); total: (n = 352); data were adjusted for female BMI in results.
Only men having a sperm concentration of at least 1 million/ml in semen; included once only in study; therefore, number of men is same as number of cycles.
Women BMI >30; FSH >10 IU.
Semen parameters; fertilization rate; number of good-quality embryos; implantation rate; positive HCG; CPR LBR.
Self-reported BMI was used. Uses delivery rate, rather than live birth rate; CPR was defined as presence of fetal heart beat on ultrasound 4–6 weeks after embryo transfer.
Schliep et al. (2015)
Prospective cohort study of men undergoing IVF–ICSI (n = 735 men)
First fresh cycles.
Men with nonobstructive azoospermia were Excluded.
Fertilization rate; embryo score (day5); CPR; LBR.
Umul et al. (2015)
Retrospective; couples (n = 55); cycles (n = 177); ICSI; significant obesity was observed in female partners of obese male BMI groups (P = 0.01). Prospective; couples (n = 163); cycles (n = 163); IVF–ICSI.
Men/cycles; same number as only first fresh cycle is included; <25, (n =224); 25–30 (n = 334); 30–35 (n = 122); > 35 (n = 55); Total (n = 735); data were adjusted for female BMI in results. Men 20–25 (n = 52) 25–30 (n = 75) > 30 (n = 28) Total couples (n = 155).
Fresh ICSI cycles with autologous oocytes.
Donor, frozen cycles, symptoms or signs of any andrological dysfunction.
Embryos transferred; fertilization rate; implantation rate; positive HCG; CPR; LBR.
CPR was defined by presence of one or more gestational sacs confirmed ultrasound, clinical recording of fetal heart tones, or both, or documentation of a birth or any abortion in case of missing ultrasound data; LBR was defined as birth in which at least one fetus was live born. GnRH antagonist. Motile sperms used in ICSI. Clinical pregnancy was defined by presence of fetal heart beat on transvaginal ultrasound.
<25 (n = 65) 25–29.9 (n = 69) > 30 (n = 29); total (n = 163); data were adjusted for female BMI in results.
Fresh cycles; autologous oocyte; day-5 elective single blastocyst transfer; fresh cycle; included once only in study; therefore, number of men is same as number of cycles. Fresh cycle; autologous oocyte; cases with high female BMI were excluded from analysis.
Donation cycles; frozen sperms.
CPR .
High male BMI does not affect CPR in IVF–ICSI cycles; CPR was defined as presence of fetal heart beat on USS 4–6 weeks after embryo transfer.
Donor Frozen cycles
CPR LBR
Consider both male and female BMIs and studied characteristics of singleton and twin newborns concieved via IVF and ICSI.
Mushtaq et al. (2016)
Wang et al. (2016)
Retrospectve cohort (n = 12,061 cycles); IVF–ICSI.
Men; <25 (n = 5305); ≥25 (n = 4417); total (n = 9722 cycles); 2339 cycles excluded (women with high BMI).
BMI, body mass index; CPR, clinical pregnancy rate; GnRH, gonadotrophin releasing hormone; ICSI, intracytoplasmic sperm injection; LBR, live birth rate.
ARTICLE IN PRESS
Studies
REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
Table 2 – (continued)
5
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
6
Figure 2 – Meta-analysis of the effect of high male body mass index on clinical pregnancy rate in assisted reproduction techniques.
Figure 3 – Meta-analysis of the effect of high male body mass index on live birth rate in assisted reproduction techniques.
Similarly, the study of Mushtaq et al. (2016) prospectively studied 160 IVF–ICSI cycles having elective single embryo transfer and concluded, after controlling for confounding factors, including female age and BMI, that raised male BMI had no significant effect on treatment outcome.
Clinical pregnancy Nine studies, (Anifandis et al., 2013; Bakos et al., 2011; Keltz et al., 2010; Merhi et al., 2013; Mushtaq et al., 2016; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015; Wang et al., 2016), including 12,294 cycles, evaluated the effect of male BMI on CPR after IVF– ICSI treatment. Pooling of the results of those studies revealed that raised male BMI was associated with a significant reduction in CPR (OR = 0.78, 95% CI 0.63 to 0.98, P = 0.03). A random-effects model was used for statistical analysis because of high statistical heterogeneity (I2 = 61%) (Figure 2).
Live birth Seven studies (Anifandis et al., 2013; Bakos et al., 2011; Petersen et al., 2013; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015; Wang et al., 2016), including 13,403 cycles, evaluated the effect of male
BMI on the LBR after IVF–ICSI treatment. Pooling of the results of those studies revealed that raised male BMI was associated with a significant reduction in LBR (OR = 0.88, 95% CI 0.82 to 0.95, P = 0.001). A fixed-effects model was used for statistical analysis because of low statistical heterogeneity (I2 = 18%) (Figure 3).
Sensitivity analysis Method of oocyte fertilization used We analysed the effect of raised male BMI on the outcome of IVF and ICSI cycles separately, where possible.
Clinical pregnancy IVF Five studies (Keltz et al., 2010; Merhi et al., 2013; Mushtaq et al., 2016; Thomsen et al., 2014; Wang et al., 2016) evaluated the effect of male BMI on CPR after IVF treatment. Pooling of the results of those studies revealed that raised male BMI was not significantly associated with CPR after IVF treatment (OR = 0.58, 95% CI 0.27 to 1.25). A randomeffects model was used for statistical analysis because of high statistical heterogeneity (I2 = 68%) (Figure 4).
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
7
Figure 4 – Meta-analysis of the effect of high male body mass index on clinical pregnancy rate in IVF and intracytoplasmic sperm injection categories separately.
ICSI Six studies (Keltz et al., 2010; Merhi et al., 2013; Mushtaq et al., 2016; Thomsen et al., 2014; Umul et al., 2015; Wang et al., 2016) evaluated the effect of male BMI on the CPR after ICSI treatment. Pooling of the results of those studies revealed that raised male BMI was not significantly associated with CPR after ICSI treatment (OR = 0.90, 95% CI 0.78 to 1.04). A fixed-effects model was used for statistical analysis because of low statistical heterogeneity (I2 = 0%) (Figure 4).
Live birth IVF Three studies (Petersen et al., 2013; Thomsen et al., 2014; Wang et al., 2016) evaluated the effect of male BMI on the LBR after IVF treatment. Pooling of the results of those studies revealed that raised male BMI was not significantly associated with LBR after IVF treatment (OR = 0.93, 95% CI 0.84 to 1.02). A fixed-effects model was used for statistical analysis because of low statistical heterogeneity (I2 = 0%) (Figure 5).
ICSI Four studies (Petersen et al., 2013; Thomsen et al., 2014; Umul et al., 2015; Wang et al., 2016) evaluated the effect of male BMI on the LBR after ICSI treatment. Pooling of the results of those studies revealed that raised male BMI was not significantly associated with LBR after ICSI treatment (OR = 0.90, 95% CI 0.77 to 1.04). A fixed-effects model was used for statistical analysis because of low statistical heterogeneity (I2 = 46%) (Figure 5).
WHO raised BMI groups We analysed the effect of raised male BMI on the outcome of IVF– ICSI cycles according to specific WHO raised BMI groups.
Clinical pregnancy Six studies (Bakos et al., 2011; Colaci et al., 2012; Mushtaq et al., 2016; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015) evaluated the effect of male BMI on CPR in the overweight and obese categories, separately. Pooling of results of those studies showed CPR was not significantly associated with BMI in both overweight (OR = 0.97, 95% CI 0.73 to 1.29) and obese (OR = 0.95, 95% CI 0.68 to 1.32) categories (Figure 6). Only two studies (Bakos et al., 2011; Schliep et al., 2015) evaluated the effect of male BMI on CPR in morbidly obese men separately. No significant association of CPR with BMI was observed in this subgroup (OR = 0.44, 95% CI 0.07 to 2.80) (Figure 6).
Live birth Six studies (Bakos et al., 2011; Colaci et al., 2012; Petersen et al., 2013; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015) evaluated the effect of male BMI on LBR in overweight and obese categories. Pooling of results of these studies revealed no significant association of LBR with BMI in the overweight subgroup (OR = 0.86, 95% CI 0.98 to 1.09); however, a statistically significant reduction in the LBR in the obese subgroup was observed (OR = 0.72, 95% CI 0.55 to 0.95, P = 0.02) (Figure 7). Only two studies (Bakos et al., 2011; Schliep et al., 2015) evaluated the effect of male BMI on the LBR in the morbidly obese subgroup.
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS 8
REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
Figure 5 – Meta-analysis of effect of high male body mass index on live birth rate in IVF and intracytoplasmic sperm injection categories separately.
Figure 6 – Meta-analysis of the effect of high male body mass index on clinical pregnancy rate according to BMI categories. Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
9
Figure 7 – Meta-analysis of effect of high male body mass index on live birth rate according to body mass index categories.
LBR was not significantly association with BMI in this subgroup (OR = 0.48, 95% CI 0.10 to 2.27) (Figure 7).
in the LBR associated with raised male BMI in the retrospective studies only (OR = 0.87, 95% CI 0.81 to 0.95, P = 0.001) (Figure 9).
Study design
Accounting for female BMI
We analysed the effect of raised male BMI on the outcome of IVF– ICSI cycles according to whether the primary studies were prospective or retrospective.
We analysed the effect of raised male BMI on the outcome of IVF– ICSI cycles according to whether the primary studies accounted for female BMI.
Clinical pregnancy Three prospective (Mushtaq et al., 2016; Schliep et al., 2015; Thomsen et al., 2014) and six retrospective studies (Anifandis et al., 2013; Bakos et al., 2011; Keltz et al., 2010; Merhi et al., 2013; Umul et al., 2015; Wang et al., 2016) evaluated the effect of male BMI on the CPR after IVF–ICSI. Pooling of results of those studies separately revealed a statistically significant reduction in the CPR associated with raised male BMI in the retrospective studies only (OR = 0.63, 95% CI 0.46 to 0.87, P = 0.005) (Figure 8).
Clinical pregnancy Seven studies (Anifandis et al., 2013; Keltz et al., 2010; Merhi et al., 2013; Mushtaq et al., 2016; Schliep et al., 2015; Thomsen et al., 2014; Wang et al., 2016) controlled for female BMI while evaluating effect of male BMI on CPR. Pooling of results of these studies revealed statistically significant reduction in CPR (OR = 0.91, 95% CI 0.84 to 0.99, P = 0.01) (Figure 10)
Live birth Live birth Three prospective (Petersen et al., 2013; Schliep et al., 2015; Thomsen et al., 2014) and four retrospective studies (Anifandis et al., 2013; Bakos et al., 2011; Umul et al., 2015; Wang et al., 2016) evaluated the effect of male BMI on the LBR after IVF–ICSI. Pooling of results of those studies separately revealed a statistically significant reduction
Five studies (Anifandis et al., 2013; Petersen et al., 2013; Schliep et al., 2015; Thomsen et al., 2014; Wang et al., 2016) either controlled or adjusted for female BMI in calculating effect of male BMI on LBR. Analysis including only these studies still indicated statistically significant reduction in LBR (OR = 0.89, 95% CI 0.82 to 0.96; P = 0.003) in men with high BMI undergoing IVF–ICSI (Figure 11).
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS 10
REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
Figure 8 – Analysis of the effect of prospective and retrospective studies on clinical pregnancy rate.
Figure 9 – Analysis of the effect of prospective and retrospective studies on live birth rate.
Discussion This systematic review and meta-analysis demonstrated that raised male BMI could have a negative effect on IVF–ICSI treatment outcome. The results showed that raised male BMI is associated with a statistically significant reduction in the clinical pregnancy and live birth rates per treatment cycle. Our study is an updated systematic review of published research. It has several strengths, including the robustness of the literature search methodology, strict study selection process which only included studies that adhered to the WHO definitions of BMI
categories, inclusion of larger number of primary studies, obtaining additional data from authors of the primary studies, and examining the influence of raised male BMI on IVF and ICSI cycles separately and in the different WHO BMI categories. Two previous systematic reviews on the subject were identified during our literature search (Campbell et al., 2015; Le et al., 2016). Both reviews included fewer studies than ours and reached conflicting conclusions. The review by Campbell et al. (2015) suggested that male obesity could reduce the pregnancy and live birth rates after IVF treatment. Unlike our study, however, this review included only five studies evaluating clinical pregnancy and live birth rates, compared normal weight with obese men and excluded overweight men
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
11
Figure 10 – Meta-analysis of the effect of high male body mass index on clinical pregnancy rate, including only studies that controlled for female body mass index.
from the analysis. The review by Le et al. (2016) concluded that raised male BMI did not affect IVF outcome. The review, however, included studies that did not report male BMI according to WHO definitions of BMI categories, rendering it difficult to group patients accurately. Although the present study demonstrated a statistically significant reduction in the CPR and LBR per IVF–ICSI treatment cycle, we urge caution in interpreting the study results. Our study identified several limitations in the available literature. The primary studies included in our systematic review were mostly retrospective and varied in the definitions used for reporting pregnancy and clinical pregnancy (Table 1) (Figure 8 and Figure 9). In addition, most studies did not account for important clinical variables influencing IVF–ICSI outcome, including female age and BMI, cause of infertility or IVF cycle characteristics, such as day of embryo transfer, number of embryos transferred and embryo quality. Furthermore, five studies (Keltz et al., 2010; Merhi et al., 2013; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015) used self-reported weight and height as opposed to standardized measurements, potentially introducing a risk of bias while allocating patients to the different BMI groups. Out of 11 studies, for the data from Umul et al. (2015), female BMI was higher than normal in the overweight and obese men categories, and indicated that it could be partly responsible for the negative effect seen on IVF outcome. The analysis did not adjust the results
for female BMI. Bakos et al. (2011) found equal distribution for female BMI in study subgroups except for morbidly obese men where female BMI was higher than normal. Anifandis et al. (2013), Petersen et al. (2013) and Wang et al. (2016) used different female BMI categories for improved sensitivity analysis, and indicated that negative effect on IVF outcomes persist despite normal BMI of female partner. Although Anifandis et al. (2013), Petersen et al. (2013) and Wang et al. (2016) demonstrated that the effect of male BMI is less pronounced than that of only high female BMI, Keltz et al. (2010) (showing effect on CPR) Colaci et al. (2012), Merhi et al. (2013), Thomsen et al. (2014), Schliep et al. (2015) (showing no effect of male BMI) and Mushtaq et al. (2016) also adjusted results for female BMI, and adjusted results are used for meta-analysis. The lack of accounting for important confounding factors could also explain the conflicting evidence in some of the published studies. For example, the large study by Kupka et al. (2010) included 700,000 assisted reproduction technique cycles from the German IVF registry, and concluded that raised male BMI did not negatively influence assisted reproduction technique outcome. That study, however, did not account for female factors, did not use the WHO definitions of BMI categories, and included overweight and normal weight men in the group, thus introducing multiple variables that could skew the study results.
Figure 11 – Meta-analysis of effect of high male body mass index on live birth rate, including only studies with controlled for female body mass index. Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
12
In addition to the methodological shortcomings identified through the literature search, pooling of the results of individual studies revealed significant statistical heterogeneity across the included studies, which necessitated using the random-effects model for statistical analysis in some of our calculations. Despite the above limitations, it seems from the study results that raised BMI is associated with a reduction in both IVF and ICSI outcomes. This suggests that ICSI may not overcome the negative effect of raised male BMI on IVF outcome, and that the negative effect of raised male BMI on treatment outcome could be mediated through a mechanism involving increased sperm DNA damage and chromatin decondensation, and impaired spermatogenesis, rather than reduction in standard WHO semen analysis parameters (Kahn and Brannigan, 2017; La Vignera et al., 2012; Zhao et al., 2014). In support of this theory, some evidence suggests that the deleterious effect of raised male BMI on sperm DNA integrity could be reversed with weight loss (Hakonsen et al., 2011). Obesity in men was noted to be associated with lower odds of live birth compared with the normal male BMI group, with non-significant differences in the other groups. Increased female BMI is also noted to affect outcomes of IVF–ICSI in the obese category (Rittenberg et al., 2011). Further studies are needed to confirm whether lesser increases in male BMI affect outcomes. The results of the sensitivity analysis should also be interpreted with caution because of the lack of statistical significance in subgroup comparisons owing to the small number of eligible studies included in the analysis of each subgroup. Therefore, it is clear from the present study that future robust prospective studies examining the relationship between male BMI and assisted reproduction technique outcome are still needed. Those studies should use standardized methods for measuring male BMI rather than relying on self-reported data, adhere to the WHO definitions of BMI categories, account for the influence of important clinical variables such as female age, female BMI, IVF and ICSI cycle characteristics including the number of embryos transferred and embryo quality, and use the live birth rate per cycle as their primary outcome. Additionally, studies evaluating the effect of weight loss and reduction of male BMI on modifying IVF–ICSI outcome are needed so that fertility specialists can provide appropriate counselling for couples embarking on treatment.
Acknowledgement We thank Mr Bolaji Coker, Senior data manager and Statistician, Guy’s and St Thomas ‘s NHS Foundation trust, London, for his suggestions in revising this article.
A R T I C L E
I N F O
Article history: Received 7 June 2017 Received in revised form 20 December 2017 Accepted 2 January 2018
Declaration: The authors report no financial or commercial conflicts of interest.
Keywords: ART Assisted reproduction Male BMI Male obesity
REFERENCES
Agarwal, A., Mulgund, A., Hamada, A., Chayatte, M.R., 2015. A unique view on male infertility around the globe. Reprod. Biol. Endocrinol. 13, 37. Andersson, A.M., Jorgensen, N., Main, K.M., Toppari, J., Rajpert-De Meyts, E., Leffers, H., Juul, A., Jensen, T.K., Skakkebaek, N.E., 2008. Adverse trends in male reproductive health: we may have reached a crucial ‘tipping point’. Int. J. Androl. Anifandis, G., Dafopoulos, K., Messini, C.I., Polyzos, N., Messinis, I.E., 2013. The BMI of men and not sperm parameters effect on embryo quality and the IVF outcome. Andrology 1, 85–89. Bakos, H.W., Henshaw, R.C., Mitchell, M., Lane, M., 2011. Paternal body mass index is associated with decreased blastocyst development and reduced live birth rates following assisted reproductive technology. Fertil. Steril. 95, 1700–1704. Braga, D.P., Halpern, G., Figueira Rde, C., Setti, A.S., Ianconelli, A., Jr., Borges, E., Jr., 2012. Food Intake and social habits in male patients and its relationship to Intracytoplasmic sperm Injection outcomes. Fertil. Steril. 97, 53–59. Campbell, J.M., Lane, M., Owens, J.A., Bakos, H.W., 2015. Paternal Obesity negatively affects male fertility and assisted reproduction outcomes: a systematic review and meta-analysis. Reprod. Biomed. Online 4, 998–1005. Centre for Reviews and Dissemination, 2001. CRD Report Number 4, 2nd edition. Undertaking systematic reviews of research on effectiveness. CRD’s guidance for carrying out or commissioning reviews. NHS Centre for Reviews and Dissemination, University of York, York, UK. Colaci, D.S., Afeiche, M., Gaskins, A.J., Wright, D.L., Toth, T.L., Tanrikut, C., Hauser, R., Chavarro, J.E., 2012. Men’s body mass index in relation to embryo quality and clinical outcomes in couples undergoing in vitro fertilization. Fertil. Steril. 98, 1193–1199. DerSimonian, R., Laird, N., 1986. Meta-analysis in clinical trials. Control. Clin. Trials 7, 177–188. Egger, M., Davey Smith, G., Schneider, M., Minder, C., 1997. Bias in meta-analysis detected by a simple graphical test. Br. Med. J. 315, 629–634. Eisenberg, M.L., Kim, S., Chen, Z., Sundaram, R., Schisterman, E.F., Buck Louis, G.M., 2014. The relationship between male BMI and waist circumference on semen quality: data from the LIFE study. Hum. Reprod. 29, 193–200. Hakonsen, L.B., Thulstrup, A.M., Aggerholm, A.S., Olsen, J., Bonde, J.P., Andersen, C.Y., Bungum, M., Ernst, E.H., Hansen, M.L., Ernst, E.H., Ramlau-Hansen, C.H., 2011. Does weight loss improve semen quality and reproductive hormones? Results from a cohort of severely obese men. Reprod. Health 8, 24. Higgins, J.P.T., Thompson, S.G., 2002. Quantifying heterogeneity in a meta-analysis. Stat. Med. 21, 1539–1558. Kahn, B.E., Brannigan, R.E., 2017. Obesity and male infertility. Curr. Opin. Urol. 27, 441–445. Katib, A., 2016. Mechanisms linking obesity to male infertility. Cent. European J. Urol. 68, 79–85. Keltz, J., Zapantis, A., Jindal, S.K., Lieman, H.J., Santoro, N., Polotsky, A.J., 2010. Overweight Men: clinical pregnancy after ART is decreased in IVF but not in ICSI cycles. J. Assist. Reprod. Genet. 27, 539–544.
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
Kupka, M.S., Gnoth, C., Buhler, K., Dahncke, W., Kruessel, J.S., 2010. Effect of female and male obesity on IVF–ICSI: results of 700,00 ART-cycles in Germany. Gynecol. Endocrinol. 0, 1–6. La Vignera, S., Condorelli, R.A., Vicari, E., Calogero, A.E., 2012. Negative effect of increased body weight on sperm conventional and nonconventional flow cytometric sperm parameters. J. Androl. 33, 53–58. Le, W., Su, S.H., Shi, L.H., Zhang, J.F., Wu, D.L., 2016. Effect of male body mass index on clinical outcomes following assisted reproductive technology: a meta-analysis in couples undergoing intracytoplasmic sperm injection. Andrologia 48, 406–424. Lewis, S., Clarke, M., 2001. Forest plots: trying to see the wood and the trees. BMJ 322, 1479–1480. Link, M.H., Schliep, K.C., Deshmukh, R., Porucznik, C.A., Carrell, D., Hammoud, A.O., 2012. Effect of paternal obesity on in vitro fertilization (IVF) pregnancy and live birth rates. Fertnstert 077, 746. Mantel, N., Haenszel, W., 1959. Statistical aspects of the analysis of data from retrospective studies of disease. J. Natl. Cancer Inst. 22, 719– 748. Merhi, Z.O., Keltz, J., Zapantis, A., Younger, J., Berger, D., Lieman, H.J., Jindal, S.K., Polotsky, A.J., 2013. Male adiposity impairs clinical pregnancy rate by In Vitro Fertilisation without affecting Day 3 embryo quality. Obesity (Silver Spring) 21, 1608–1612. Mushtaq, R., El-Toukhy, T., Khalaf, Y., Gebhardt, R., Rittenberg, V. 2016. Effect of male BMI on IVF outcome after elective single blastocyst transfer. A prospective study. Poster presented at European Society of Human Reproduction and Endocrinology annual meeting; July 2016 3–6.Helsinki, Finland. National Institute for Health and Care Excellence, 2013. Fertility Problems: Assessment and Treatment. NICE, London. Nguyen, D.M., El-Serag, H.B., 2010. The epidemiology of obesity. Gastroenterology 39, 1–7. Petersen, G.L., Schmidt, L., Pinborg, A., Kamper-Jorgensen, M., 2013. The influence of female and male body mass index on live births after assisted reproductive technology treatment: a nationwide register-based cohort study. Fertil. Steril. 99, 1654–1662. Popkin, B.M., Adair, L.S., Ng, S.W., 2012. Global nutrition transition and pandemic of obesity in developing countries. Nutr. Rev. 70, 3–21. Rittenberg, V., Seshadri, S., Sunkara, S.K., Sobaleva, S., Oteng-Ntim, E., El-Toukhy, T., 2011. Effect of body mass index on IVF treatment outcome: an updated systematic review and meta-analysis. Reprod. Biomed. Online 23, 421–439. Schliep, K.C., Mumford, S.L., Ahrens, K.A., Hotaling, J.M., Carrell, D.T., Link, M., Hinkle, S.N., Kissell, K., Porucznik, C.A.,
13
Hammoud, A.O., 2015. Effect of male and female body mass index on pregnancy and live birth success after in vitro fertilization. Fertil. Steril. 103, 388–395. Stroup, D., Berlin, J., Morton, S., Olkin, I., Williamson, G.D., Rennie, D., Moher, D., Becker, B.J., Sipe, T.A., Thacker, S.B., 2000. Metaanalysis of observational studies in epidemiology (MOOSE) group. J. Am. Med. Assoc. 283, 2008–2012. Sundaram, R., Sunni, L., Germaine, M., Buck, L., 2017. Couples’ body composition and time-to-pregnancy. Hum. Reprod. 32, 662–668. Thomsen, L., Humaidan, P., Bungum, L., Bungum, M., 2014. The effect of male overweight on semen quality and outcome of assisted reproduction. Asian J. Androl. 16, 749–754. Thonneau, P., Marchand, S., Tallec, A., Ferial, M.L., Ducot, B., Lansac, J., Lopes, P., Tabaste, J.M., Spira, A., 1991. Incidence and main causes of infertility in a resident population (1,850,000) of three French regions (1988–1989). Hum. Reprod. 6, 811–816. Tsao, C.W., Liu, C.Y., Chou, Y.C., Cha, T.L., Chen, S.C., Hsu, C.Y., 2015. Exploration of the association between obesity and semen quality in a 7630 male population. PLoS ONE 10. Umul, M., Kose, S.A., Bilen, E., Altuncu, A.G., Oksay, T., Guney, M., 2015. Effect of increasing paternal body mass index on pregnancy and live birth rates in couples undergoing intracytoplasmic sperm injection. Andrologia 47, 360–364. Wang, X., Hao, J., Zhang, F., Li, J., Kong, H., Guo, Y., 2016. Effect of female and male body mass indices on the treatment outcomes and neonatal birthweights associated with in vitro fertilization/ intracytoplasmic sperm injection treatment in China. Fertil. Steril. 106, 2–460. Wells, G., Shea, B., O’Connell, D. 2000. The Newcastle-Ottawa Scale (NOS) for assessing the quality of non-randomized studies in meta-analysis. Proceedings of the Third Symposium on Systematic Reviews. Beyond the Basics: improving Quality and Impact. 3—5 July, Oxford. WHO, 2017. Obesity and Overweight. World Health Organization, Geneva. http://www.who.int/mediacentre/factsheets/fs311/en/. Fact Sheet 311. Wu, Z., Lu, X., Wang, M., Cheng, H., 2015. Correlation between body mass index of chinese males and assisted reproductive technology outcome. Int. J. Clin. Exp. Med. 8, 21472. Zhao, J., Zhang, Q., Wang, Y., Li, Y., 2014. Whether sperm deoxyribonucleic acid fragmentation has an effect on pregnancy and miscarriage after in vitro fertilization/intracytoplasmic sperm injection: a systematic review and meta-analysis. Reprod. Biomed. Online 4, 998–1005.
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
Review
Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis Rabia Mushtaq a,*, Jyotsna Pundir b, Chiara Achilli c, Osama Naji a, Yacoub Khalaf a, Tarek El-Toukhy a a Assisted Conception Unit, Guy’s and St Thomas Hospital NHS Foundation Trust, 11th Floor, Tower Wing, London, UK b Centre of Reproductive Medicine, St Bartholomew’s Hospital, London, UK c The Hewitt Fertility Centre, Liverpool Women’s Hospital, Liverpool, UK
Ms Rabia Mushtaq (MBBS, FCPS, MRCOG) currently works at Milton Keynes University Hospital, and developed an interest in research while working as Clinical Research Fellow at Guy’s and St Thomas Foundation Trust in assisted conception. Her special interests are factors influencing IVF and pregnancy outcomes. KEY MESSAGE Published research suggests that raised body mass index (BMI) has a negative effect on IVF and intracytoplasmic sperm injection treatment outcome. Future well-designed, robust prospective studies adhering to the World Health Organization definitions of BMI categories and considering important confounding variables are needed to confirm our study results.
A B S T R A C T Men with a body mass index (BMI) of 30 or over are more likely to have reduced fertility and fecundity rates. This systematic review and meta-analysis evaluated the effect of male BMI on IVF and intracytoplasmic sperm injection (ICSI) outcome. An electronic search for published literature was conducted in MEDLINE and EMBASE between 1966 and November 2016. Outcome measures were clinical pregnancy rates (CPR) and live birth rates (LBR) per IVF or ICSI cycle. Eleven studies were identified, including 14,372 cycles; nine reported CPR and seven reported LBR. Pooling of data from those studies revealed that raised male BMI was associated with a significant reduction in CPR (OR 0.78, 95% CI 0.63 to 0.98, P = 0.03) and LBR (OR 0.88, 95% CI 0.82 to 0.95, P = 0.001) per IVF–ICSI treatment cycle. Male BMI could be an important factor influencing IVF–ICSI outcome. More robust studies are needed to confirm this conclusion using standardized methods for measuring male BMI, adhering to the World Health Organization definitions of BMI categories, accounting for female BMI, IVF and ICSI cycle characteristics, including the number of embryos transferred and embryo quality, and use the live birth rate per cycle as primary outcome. © 2018 Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. E-mail address: [email protected] (R Mushtaq). https://doi.org/10.1016/j.rbmo.2018.01.002 1472-6483/© 2018 Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
2
Introduction Obesity is a global pandemic, which carries major health consequences and negative effect on quality of life. Worldwide obesity has more than doubled since 1980 (Nguyen and El-Serag, 2010), and is projected to increase further owing to anticipated demographic changes (Popkin et al, 2012). Male factors are thought to contribute to up to 50% of subfertility cases, with 31.5% being attributed solely to a male factor cause (Agarwal et al., 2015; Thonneau et al., 1991). Men who have a body mass index (BMI, measured as kg/m2) of 30 or more are more likely to have reduced fertility and fecundity rates (NICE, 2013; Sundaram et al., 2017). Several mechanisms have been attributed to the reduced fecundity in obese men, including lower serum testosterone and raised serum oestradiol levels, impaired spermatogenesis and erectile dysfunction (Andersson et al., 2008; Katib, 2016). Recent data evaluating the relationship between waist circumference and semen parameters in men without known infertility, indicated higher prevalence of lower ejaculatory volume and oligozoospermia in men with increasing waist circumference and BMI (Eisenberg et al., 2014). In addition, lower sperm concentration and higher percentage of abnormal sperm morphology were associated with increasing body adiposity in a recent cross-sectional cohort study (Tsao et al., 2015). Evidence also shows that raised male BMI could be associated with a lower success rate after IVF treatment. A large cohort study showed that couples in whom both partners were either overweight or obese had the lowest odds for live birth after IVF (Petersen et al., 2013). Furthermore, a large observational study demonstrated that male partner BMI had a greater effect on embryo quality and IVF outcome than semen analysis parameters (Anifandis et al., 2013). Increased level of sperm DNA damage associated with male obesity has been shown to be related to lower pregnancy and higher miscarriage rates in both IVF and ICSI cycles (Zhao et al., 2014). The results of published studies addressing the relationship between raised male BMI and IVF–ICSI outcome, however, are conflicting, with some studies reporting a negative effect of raised male BMI (Bakos et al., 2011; Merhi et al., 2013; Umul et al., 2015), and others reporting no effect (Braga et al., 2012; Schliep et al., 2015; Thomsen et al., 2014). In this study, we sought to systematically review and summarize the existing evidence related to the effect of male BMI on clinical pregnancy and live birth rates after IVF–ICSI treatment.
Materials and methods Literature search methodology Electronic searches for published literature in MEDLINE, and EMBASE were conducted from database inception until November 2016 to capture citations including male partner BMI and reproductive outcome after IVF–ICSI treatment. A combination of medical subject headings (MeSH) and text words were used to generate two subsets of citations: one including ‘male body mass index’ (paternal obesity, BMI, paternal body mass index, male adiposity, overweight men) and the second subset for assisted reproductive techniques (assisted reproductive technology, assisted reproduction, ART cycles, IVF, ICSI, in
vitro fertilisation, invitro fertilisation, in-vitro fertilization, invitro fertilization and intra-cytoplasmic sperm injection). These subsets were combined with ‘AND’ to generate a subset of citations relevant to our research question. No language restrictions were applied. The reference lists of all known primary and review articles were examined to identify relevant articles not captured by electronic searches.
Study selection and outcome measures We considered prospective and retrospective cohort studies, published in full or as abstracts, examining the effect of male BMI on the outcome of IVF and ICSI treatment. Participants were male partners of couples undergoing IVF or ICSI, for whom information about male BMI was available and who were using ejaculated sperm for IVF or ICSI treatment. Studies involving natural conceptions, oocyte donation cycles, intrauterine insemination, ovulation induction and those not using the World Health Organization (WHO) criteria for BMI reporting were excluded. Studies were selected in a two-stage process. In the first instance, two reviewers (RM and CA) independently scrutinized all the titles and abstracts from the electronic searches and full manuscripts of all citations that definitely, or possibly, met the predefined selection criteria were retrieved. After examining the full manuscripts, final inclusion or exclusion decisions were made. Any disagreement about inclusion was resolved by consensus after consultation with a third reviewer (JP). The outcome measures considered for this review were clinical pregnancy and live birth rates per cycle started.
Data extraction and quality assessment Data extraction and quality assessment was carried out independently by two reviewers (RM and CA). For each study included, information was obtained on population size, study design, male BMI categories used, number of participants in each category and study outcome measures. Duplicate studies or studies with overlapping populations were excluded. In cases of missing or unclear data, the authors of the primary studies were contacted. The selected studies were assessed for methodological quality by using the components of study design that are related to internal validity (Centre for Reviews and Dissemination, 2001). Meta-analysis of observational studies in epidemiology (MOOSE) guidelines were followed (Stroup et al., 2000). The methodological quality and risk of bias of each study was assessed using the Newcastle–Ottawa quality assessment scale (NOS) (Wells et al., 2000). A study with NOS score of 6 or higher was regarded as a high-quality study. We combined data from studies if they had similar design, intervention and outcome measures. Data collected compared the clinical pregnancy rate (CPR) and live birth rate (LBR) in the study group (male BMI of 25 or more, i.e. overweight and obese men) with a control group of men with a normal BMI (18.5–24.9). A sensitivity analysis on overweight (BMI 25–29.9), obese (BMI 30–34.9) and morbidly obese (BMI ≥35) men was carried out, and each subgroup was compared with the normal BMI group (BMI 18.5–24.9) for both outcome measures. Similarly, studies that separated data for IVF and ICSI cycles were analysed in another subgroup analysis for the CPR and LBR. A further analysis was carried out including only studies that accounted for female BMI.
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
Statistical analysis From each study, binary data were extracted in 2 × 2 tables, and the results were pooled and summarized as relative risks (RR) with 95% confidence interval (CI) using fixed-effects (Mantel and Haenszel, 1959) or random-effects model as appropriate (DerSimonian and Laird, 1986). Heterogeneity of the exposure effects was evaluated graphically using forest plots (Lewis and Clarke, 2001) and statistically using the I2 statistic to quantify heterogeneity across studies (Higgins and Thompson, 2002). Substantial heterogeneity was considered if the I2 was more than 50% and considered when interpreting the results. Clinical heterogeneity was analysed based on the variation in features of the population, intervention and study quality. RevMan 5.3 software (The Cochrane Collaboration, Oxford, UK) was used for all statistical analyses. To assess for publication bias, a funnel plot analysis using the Egger test was conducted (Egger et al., 1997).
Results Study identification and selection is summarized in Figure 1. The search strategy yielded 92 citations captured from the electronic search, and 20 were obtained from review of reference list of original and review articles. Of these 112 studies, 19 duplicates were removed. Of the remaining 93 studies, 78 were excluded as it was clear from the title or abstract that they did not fulfil the selection criteria. Full manuscripts were retrieved for the remaining 15 studies for detailed evaluation. Four articles (Braga et al., 2012; Kupka et al., 2010; Link et al., 2012; Wu et al., 2015) were then excluded (Table 1) and
Figure 1 – Study identification and selection (11 articles included in systematic review).
3
Table 1 – Excluded studies. Study
Reason
Kupka et al. (2010)
The comparison was only between BMI under 30 and over 30. As both normal weight and overweight men were categorized together, data could not be used. Linear data, interpreted as increasing BMI, rather than WHO BMI categories (WHO, 2017). Definition of normal weight, overweight and obese was not in accordance with WHO criteria. Abstract only, more robust data inclusive of the period covered by this study was included by Schliep et al. (2015). This study is not included in the analysis, because of the risk of duplicating data.
Braga et al. (2012) Wu et al. (2015) Link et al. (2012)
BMI, body mass index; WHO, World Health Organization.
11 studies (Anifandis et al., 2013; Bakos et al., 2011; Colaci et al., 2012; Keltz et al., 2010; Merhi et al., 2013; Mushtaq et al., 2016; Petersen et al., 2013; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015; Wang et al., 2016) were included in this systematic review and meta-analysis (Table 2).
Study characteristics The 11 studies selected in our review included 14,372 IVF–ICSI cycles. Of those 11 studies, seven reported both CPR and LBR (Anifandis et al., 2013; Bakos et al., 2011; Colaci et al., 2012; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015; Wang et al., 2016). Three studies, however, reported only the CPR (Keltz et al., 2010, Merhi et al., 2013, and Mushtaq et al., 2016) and one study reported only the LBR (Petersen et al., 2013). The characteristics of the 11 selected trials are presented in Table 2. Keltz et al. (2010) retrospectively assessed the CPR in 290 IVF– ICSI cycles, and concluded that raised male BMI adversely affected the CPR after IVF, but not ICSI, cycles. Bakos et al. (2011) conducted a retrospective analysis of 305 fresh IVF–ICSI cycles, and reported that raised male BMI was associated with poorer blastocyst development and reduced CPR and LBR. Colaci et al. (2012) prospectively studied 172 cycles in 114 couples and concluded that raised male BMI had no effect on IVF–ICSI outcome. The study stated outcomes in IVF and ICSI categories separately as well as in BMI categories, and expressed it as odds ratio rather than numbers for each category. Therefore, data from this study was included only in subgroup analysis and not included in main groups, i.e, CPR and LBR in total. The retrospective studies of Anifandis et al. (2013) and Merhi et al. (2013) included 301 and 344 cycles respectively, and both concluded that raised male BMI negatively influenced pregnancy rates after IVF–ICSI cycles, although they reported different effects on embryo quality. Petersen et al. (2013) reported a population-based cohort study of 25,191 IVF–ICSI cycles and concluded the raised male BMI negatively influenced IVF and, to a lesser extent, ICSI outcome. Petersen et al. (2013) and Anifandis et al. (2013) presented data in four group categories. Only groups with female BMI less than 25 were included in the analysis. Similarly, Umul et al. (2015) retrospectively analysed 177 ICSI cycles and reported that raised male BMI was associated with lower implantation, pregnancy and live birth rates. In contrast, Thomsen et al. (2014) and Schliep et al. (2015) prospectively studied 612 and 735 fresh IVF–ICSI cycles, respectively, and concluded that raised male BMI had no effect on IVF–ICSI outcome.
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
4
Study type and number of participants/cycles
BMI group and numbers included in analysis
Inclusion criteria
Exclusion criteria
Outcome measures
Comments
Keltz et al. (2010)
Retrospective Cycles (n = 290); IVF–ICSI.
Fresh cycles.
Donor Frozen cycles.
Semen parameters; day-3 embryos; numbers; implantation rate; CPR.
CPR was defined as intrauterine gestational sac on transvaginal ultrasound; number of couples not mentioned.
Bakos et al. (2011)
Retrospective men (n = 305); IVF–ICSI; no difference in maternal BMI for male grouping of normal, overweight, or obese. In morbidly obese male BMI group, maternal BMI was higher compared with normal-weight group (P < 0.01). Prospective cycles (n = 172) Couples undergoing IVF–ICSI (n = 114).
<25 (n = 62 cycles); ≥ 25 (n = 228 cycles) Total (n = 290 cycles); data were adjusted for female BMI in results. Men/cycles < 25 (n = 63); 25–30 (n = 148); 30–35 (n = 62); > 35 (n = 32); total (n = 305).
Female partner aged <38 years; fresh cycle included once only in study; therefore, number of men is same as number of cycles.
Donor or frozen sperm treatment; andrological dysfunction.
Semen parameters; fertilization rate; embryo quality; implantation rate; positive HCG; CPR; LBR; miscarriage rate.
GnRH antagonist protocol. Clinical pregnancy indicated by the presence of a viable fetal heart beat on ultrasound 4–6 weeks after embryo transfer.
More than one cycle/couple included.
Donor.
Semen parameters; fertilization rate; embryo quality ; implantation rate; CPR; LBR.
Fresh cycle included once only in study; therefore, number of men is same as number of cycles.
Not mentioned.
Semen parameters; embryo quality; implantation rate; positive HCG; CPR; LBR.
Zygote pronuclear score; day-3 embryo numbers; day-3 grade for transferred embryos; CPR. Live birth.
CPR was defined as presence of intrauterine pregnancy confirmed by ultrasound; LBR was defined as birth of a neonate after 24 weeks of gestation. Clinical pregnancy indicated by intrauterine sac seen by ultrasound 3–4 weeks after HCG; subgroup analysis for overweight and obese was not included, neither was analysis for IVF or ICSI; day of embryo transfer not documented; BMI of female partner was controlled for. All sperm injected into oocytes were motile.
Colaci et al. (2012)
<25 (n = 54 cycles); 25–30 (n = 80) cycles > 30 (n = 38) cycles Total (n = 172); data were adjusted for female BMI in results. Men BMI >25 and women normal BMI, (n = 142 couples); both men and women with normal BMI (n = 64); total (n = 206).
Anifandis et al. (2013)
Total retrospective men undergoing IVF–ICSI (n = 301).
Merhi et al. (2013)
Retrospective cycles of IVF–ICSI (n = 344)
<25 (n = 75 cycles); ≥25 (n = 269 cycles); total (n = 344 cycles); data were adjusted for female BMI in results.
Fresh cycles.
Donor Frozen cycles.
Petersen et al. (2013)
Population-based cohort study; couples (n = 12566); IVF–ICSI cycles for which male BMI information was available (n = 1906).
Men and women < 18.5 (n = 9); cycles (not included in Figure 7 analysis); 18.5–25 (n = 905 cycles); > 25–30 (n = 742 cycles); > 30 (n = 250 cycles); total (n = 1906 Only cycles where female BMI was normal were included in analysis in Figures 3,5,9 and 11.
Autologous oocytes.
Donor.
Increased BMI in either male or female negatively influence live birth rate, and this association is less clear with ICSI.
(continued on next page)
ARTICLE IN PRESS
Studies
REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
Table 2 – Characteristics of included studies.
Study type and number of participants/cycles
BMI group and numbers included in analysis
Inclusion criteria
Exclusion criteria
Outcome measures
Comments
Thomsen et al. (2014)
Prospective; (n = 612 couples in total); 86 split cycle (IVF + ICSI); 167 IUI were excluded from analysis; IVF, 233 started IVF and 126 ICSI. Couples in final analysis who had oocyte retrieval 228 + 124 ICSI =352 in total.
< 25 (n = 148 men); 25–30 (n = 160 men); > 30 (n = 44 men); total: (n = 352); data were adjusted for female BMI in results.
Only men having a sperm concentration of at least 1 million/ml in semen; included once only in study; therefore, number of men is same as number of cycles.
Women BMI >30; FSH >10 IU.
Semen parameters; fertilization rate; number of good-quality embryos; implantation rate; positive HCG; CPR LBR.
Self-reported BMI was used. Uses delivery rate, rather than live birth rate; CPR was defined as presence of fetal heart beat on ultrasound 4–6 weeks after embryo transfer.
Schliep et al. (2015)
Prospective cohort study of men undergoing IVF–ICSI (n = 735 men)
First fresh cycles.
Men with nonobstructive azoospermia were Excluded.
Fertilization rate; embryo score (day5); CPR; LBR.
Umul et al. (2015)
Retrospective; couples (n = 55); cycles (n = 177); ICSI; significant obesity was observed in female partners of obese male BMI groups (P = 0.01). Prospective; couples (n = 163); cycles (n = 163); IVF–ICSI.
Men/cycles; same number as only first fresh cycle is included; <25, (n =224); 25–30 (n = 334); 30–35 (n = 122); > 35 (n = 55); Total (n = 735); data were adjusted for female BMI in results. Men 20–25 (n = 52) 25–30 (n = 75) > 30 (n = 28) Total couples (n = 155).
Fresh ICSI cycles with autologous oocytes.
Donor, frozen cycles, symptoms or signs of any andrological dysfunction.
Embryos transferred; fertilization rate; implantation rate; positive HCG; CPR; LBR.
CPR was defined by presence of one or more gestational sacs confirmed ultrasound, clinical recording of fetal heart tones, or both, or documentation of a birth or any abortion in case of missing ultrasound data; LBR was defined as birth in which at least one fetus was live born. GnRH antagonist. Motile sperms used in ICSI. Clinical pregnancy was defined by presence of fetal heart beat on transvaginal ultrasound.
<25 (n = 65) 25–29.9 (n = 69) > 30 (n = 29); total (n = 163); data were adjusted for female BMI in results.
Fresh cycles; autologous oocyte; day-5 elective single blastocyst transfer; fresh cycle; included once only in study; therefore, number of men is same as number of cycles. Fresh cycle; autologous oocyte; cases with high female BMI were excluded from analysis.
Donation cycles; frozen sperms.
CPR .
High male BMI does not affect CPR in IVF–ICSI cycles; CPR was defined as presence of fetal heart beat on USS 4–6 weeks after embryo transfer.
Donor Frozen cycles
CPR LBR
Consider both male and female BMIs and studied characteristics of singleton and twin newborns concieved via IVF and ICSI.
Mushtaq et al. (2016)
Wang et al. (2016)
Retrospectve cohort (n = 12,061 cycles); IVF–ICSI.
Men; <25 (n = 5305); ≥25 (n = 4417); total (n = 9722 cycles); 2339 cycles excluded (women with high BMI).
BMI, body mass index; CPR, clinical pregnancy rate; GnRH, gonadotrophin releasing hormone; ICSI, intracytoplasmic sperm injection; LBR, live birth rate.
ARTICLE IN PRESS
Studies
REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
Table 2 – (continued)
5
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
6
Figure 2 – Meta-analysis of the effect of high male body mass index on clinical pregnancy rate in assisted reproduction techniques.
Figure 3 – Meta-analysis of the effect of high male body mass index on live birth rate in assisted reproduction techniques.
Similarly, the study of Mushtaq et al. (2016) prospectively studied 160 IVF–ICSI cycles having elective single embryo transfer and concluded, after controlling for confounding factors, including female age and BMI, that raised male BMI had no significant effect on treatment outcome.
Clinical pregnancy Nine studies, (Anifandis et al., 2013; Bakos et al., 2011; Keltz et al., 2010; Merhi et al., 2013; Mushtaq et al., 2016; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015; Wang et al., 2016), including 12,294 cycles, evaluated the effect of male BMI on CPR after IVF– ICSI treatment. Pooling of the results of those studies revealed that raised male BMI was associated with a significant reduction in CPR (OR = 0.78, 95% CI 0.63 to 0.98, P = 0.03). A random-effects model was used for statistical analysis because of high statistical heterogeneity (I2 = 61%) (Figure 2).
Live birth Seven studies (Anifandis et al., 2013; Bakos et al., 2011; Petersen et al., 2013; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015; Wang et al., 2016), including 13,403 cycles, evaluated the effect of male
BMI on the LBR after IVF–ICSI treatment. Pooling of the results of those studies revealed that raised male BMI was associated with a significant reduction in LBR (OR = 0.88, 95% CI 0.82 to 0.95, P = 0.001). A fixed-effects model was used for statistical analysis because of low statistical heterogeneity (I2 = 18%) (Figure 3).
Sensitivity analysis Method of oocyte fertilization used We analysed the effect of raised male BMI on the outcome of IVF and ICSI cycles separately, where possible.
Clinical pregnancy IVF Five studies (Keltz et al., 2010; Merhi et al., 2013; Mushtaq et al., 2016; Thomsen et al., 2014; Wang et al., 2016) evaluated the effect of male BMI on CPR after IVF treatment. Pooling of the results of those studies revealed that raised male BMI was not significantly associated with CPR after IVF treatment (OR = 0.58, 95% CI 0.27 to 1.25). A randomeffects model was used for statistical analysis because of high statistical heterogeneity (I2 = 68%) (Figure 4).
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
7
Figure 4 – Meta-analysis of the effect of high male body mass index on clinical pregnancy rate in IVF and intracytoplasmic sperm injection categories separately.
ICSI Six studies (Keltz et al., 2010; Merhi et al., 2013; Mushtaq et al., 2016; Thomsen et al., 2014; Umul et al., 2015; Wang et al., 2016) evaluated the effect of male BMI on the CPR after ICSI treatment. Pooling of the results of those studies revealed that raised male BMI was not significantly associated with CPR after ICSI treatment (OR = 0.90, 95% CI 0.78 to 1.04). A fixed-effects model was used for statistical analysis because of low statistical heterogeneity (I2 = 0%) (Figure 4).
Live birth IVF Three studies (Petersen et al., 2013; Thomsen et al., 2014; Wang et al., 2016) evaluated the effect of male BMI on the LBR after IVF treatment. Pooling of the results of those studies revealed that raised male BMI was not significantly associated with LBR after IVF treatment (OR = 0.93, 95% CI 0.84 to 1.02). A fixed-effects model was used for statistical analysis because of low statistical heterogeneity (I2 = 0%) (Figure 5).
ICSI Four studies (Petersen et al., 2013; Thomsen et al., 2014; Umul et al., 2015; Wang et al., 2016) evaluated the effect of male BMI on the LBR after ICSI treatment. Pooling of the results of those studies revealed that raised male BMI was not significantly associated with LBR after ICSI treatment (OR = 0.90, 95% CI 0.77 to 1.04). A fixed-effects model was used for statistical analysis because of low statistical heterogeneity (I2 = 46%) (Figure 5).
WHO raised BMI groups We analysed the effect of raised male BMI on the outcome of IVF– ICSI cycles according to specific WHO raised BMI groups.
Clinical pregnancy Six studies (Bakos et al., 2011; Colaci et al., 2012; Mushtaq et al., 2016; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015) evaluated the effect of male BMI on CPR in the overweight and obese categories, separately. Pooling of results of those studies showed CPR was not significantly associated with BMI in both overweight (OR = 0.97, 95% CI 0.73 to 1.29) and obese (OR = 0.95, 95% CI 0.68 to 1.32) categories (Figure 6). Only two studies (Bakos et al., 2011; Schliep et al., 2015) evaluated the effect of male BMI on CPR in morbidly obese men separately. No significant association of CPR with BMI was observed in this subgroup (OR = 0.44, 95% CI 0.07 to 2.80) (Figure 6).
Live birth Six studies (Bakos et al., 2011; Colaci et al., 2012; Petersen et al., 2013; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015) evaluated the effect of male BMI on LBR in overweight and obese categories. Pooling of results of these studies revealed no significant association of LBR with BMI in the overweight subgroup (OR = 0.86, 95% CI 0.98 to 1.09); however, a statistically significant reduction in the LBR in the obese subgroup was observed (OR = 0.72, 95% CI 0.55 to 0.95, P = 0.02) (Figure 7). Only two studies (Bakos et al., 2011; Schliep et al., 2015) evaluated the effect of male BMI on the LBR in the morbidly obese subgroup.
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS 8
REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
Figure 5 – Meta-analysis of effect of high male body mass index on live birth rate in IVF and intracytoplasmic sperm injection categories separately.
Figure 6 – Meta-analysis of the effect of high male body mass index on clinical pregnancy rate according to BMI categories. Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
9
Figure 7 – Meta-analysis of effect of high male body mass index on live birth rate according to body mass index categories.
LBR was not significantly association with BMI in this subgroup (OR = 0.48, 95% CI 0.10 to 2.27) (Figure 7).
in the LBR associated with raised male BMI in the retrospective studies only (OR = 0.87, 95% CI 0.81 to 0.95, P = 0.001) (Figure 9).
Study design
Accounting for female BMI
We analysed the effect of raised male BMI on the outcome of IVF– ICSI cycles according to whether the primary studies were prospective or retrospective.
We analysed the effect of raised male BMI on the outcome of IVF– ICSI cycles according to whether the primary studies accounted for female BMI.
Clinical pregnancy Three prospective (Mushtaq et al., 2016; Schliep et al., 2015; Thomsen et al., 2014) and six retrospective studies (Anifandis et al., 2013; Bakos et al., 2011; Keltz et al., 2010; Merhi et al., 2013; Umul et al., 2015; Wang et al., 2016) evaluated the effect of male BMI on the CPR after IVF–ICSI. Pooling of results of those studies separately revealed a statistically significant reduction in the CPR associated with raised male BMI in the retrospective studies only (OR = 0.63, 95% CI 0.46 to 0.87, P = 0.005) (Figure 8).
Clinical pregnancy Seven studies (Anifandis et al., 2013; Keltz et al., 2010; Merhi et al., 2013; Mushtaq et al., 2016; Schliep et al., 2015; Thomsen et al., 2014; Wang et al., 2016) controlled for female BMI while evaluating effect of male BMI on CPR. Pooling of results of these studies revealed statistically significant reduction in CPR (OR = 0.91, 95% CI 0.84 to 0.99, P = 0.01) (Figure 10)
Live birth Live birth Three prospective (Petersen et al., 2013; Schliep et al., 2015; Thomsen et al., 2014) and four retrospective studies (Anifandis et al., 2013; Bakos et al., 2011; Umul et al., 2015; Wang et al., 2016) evaluated the effect of male BMI on the LBR after IVF–ICSI. Pooling of results of those studies separately revealed a statistically significant reduction
Five studies (Anifandis et al., 2013; Petersen et al., 2013; Schliep et al., 2015; Thomsen et al., 2014; Wang et al., 2016) either controlled or adjusted for female BMI in calculating effect of male BMI on LBR. Analysis including only these studies still indicated statistically significant reduction in LBR (OR = 0.89, 95% CI 0.82 to 0.96; P = 0.003) in men with high BMI undergoing IVF–ICSI (Figure 11).
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS 10
REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
Figure 8 – Analysis of the effect of prospective and retrospective studies on clinical pregnancy rate.
Figure 9 – Analysis of the effect of prospective and retrospective studies on live birth rate.
Discussion This systematic review and meta-analysis demonstrated that raised male BMI could have a negative effect on IVF–ICSI treatment outcome. The results showed that raised male BMI is associated with a statistically significant reduction in the clinical pregnancy and live birth rates per treatment cycle. Our study is an updated systematic review of published research. It has several strengths, including the robustness of the literature search methodology, strict study selection process which only included studies that adhered to the WHO definitions of BMI
categories, inclusion of larger number of primary studies, obtaining additional data from authors of the primary studies, and examining the influence of raised male BMI on IVF and ICSI cycles separately and in the different WHO BMI categories. Two previous systematic reviews on the subject were identified during our literature search (Campbell et al., 2015; Le et al., 2016). Both reviews included fewer studies than ours and reached conflicting conclusions. The review by Campbell et al. (2015) suggested that male obesity could reduce the pregnancy and live birth rates after IVF treatment. Unlike our study, however, this review included only five studies evaluating clinical pregnancy and live birth rates, compared normal weight with obese men and excluded overweight men
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
11
Figure 10 – Meta-analysis of the effect of high male body mass index on clinical pregnancy rate, including only studies that controlled for female body mass index.
from the analysis. The review by Le et al. (2016) concluded that raised male BMI did not affect IVF outcome. The review, however, included studies that did not report male BMI according to WHO definitions of BMI categories, rendering it difficult to group patients accurately. Although the present study demonstrated a statistically significant reduction in the CPR and LBR per IVF–ICSI treatment cycle, we urge caution in interpreting the study results. Our study identified several limitations in the available literature. The primary studies included in our systematic review were mostly retrospective and varied in the definitions used for reporting pregnancy and clinical pregnancy (Table 1) (Figure 8 and Figure 9). In addition, most studies did not account for important clinical variables influencing IVF–ICSI outcome, including female age and BMI, cause of infertility or IVF cycle characteristics, such as day of embryo transfer, number of embryos transferred and embryo quality. Furthermore, five studies (Keltz et al., 2010; Merhi et al., 2013; Schliep et al., 2015; Thomsen et al., 2014; Umul et al., 2015) used self-reported weight and height as opposed to standardized measurements, potentially introducing a risk of bias while allocating patients to the different BMI groups. Out of 11 studies, for the data from Umul et al. (2015), female BMI was higher than normal in the overweight and obese men categories, and indicated that it could be partly responsible for the negative effect seen on IVF outcome. The analysis did not adjust the results
for female BMI. Bakos et al. (2011) found equal distribution for female BMI in study subgroups except for morbidly obese men where female BMI was higher than normal. Anifandis et al. (2013), Petersen et al. (2013) and Wang et al. (2016) used different female BMI categories for improved sensitivity analysis, and indicated that negative effect on IVF outcomes persist despite normal BMI of female partner. Although Anifandis et al. (2013), Petersen et al. (2013) and Wang et al. (2016) demonstrated that the effect of male BMI is less pronounced than that of only high female BMI, Keltz et al. (2010) (showing effect on CPR) Colaci et al. (2012), Merhi et al. (2013), Thomsen et al. (2014), Schliep et al. (2015) (showing no effect of male BMI) and Mushtaq et al. (2016) also adjusted results for female BMI, and adjusted results are used for meta-analysis. The lack of accounting for important confounding factors could also explain the conflicting evidence in some of the published studies. For example, the large study by Kupka et al. (2010) included 700,000 assisted reproduction technique cycles from the German IVF registry, and concluded that raised male BMI did not negatively influence assisted reproduction technique outcome. That study, however, did not account for female factors, did not use the WHO definitions of BMI categories, and included overweight and normal weight men in the group, thus introducing multiple variables that could skew the study results.
Figure 11 – Meta-analysis of effect of high male body mass index on live birth rate, including only studies with controlled for female body mass index. Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
12
In addition to the methodological shortcomings identified through the literature search, pooling of the results of individual studies revealed significant statistical heterogeneity across the included studies, which necessitated using the random-effects model for statistical analysis in some of our calculations. Despite the above limitations, it seems from the study results that raised BMI is associated with a reduction in both IVF and ICSI outcomes. This suggests that ICSI may not overcome the negative effect of raised male BMI on IVF outcome, and that the negative effect of raised male BMI on treatment outcome could be mediated through a mechanism involving increased sperm DNA damage and chromatin decondensation, and impaired spermatogenesis, rather than reduction in standard WHO semen analysis parameters (Kahn and Brannigan, 2017; La Vignera et al., 2012; Zhao et al., 2014). In support of this theory, some evidence suggests that the deleterious effect of raised male BMI on sperm DNA integrity could be reversed with weight loss (Hakonsen et al., 2011). Obesity in men was noted to be associated with lower odds of live birth compared with the normal male BMI group, with non-significant differences in the other groups. Increased female BMI is also noted to affect outcomes of IVF–ICSI in the obese category (Rittenberg et al., 2011). Further studies are needed to confirm whether lesser increases in male BMI affect outcomes. The results of the sensitivity analysis should also be interpreted with caution because of the lack of statistical significance in subgroup comparisons owing to the small number of eligible studies included in the analysis of each subgroup. Therefore, it is clear from the present study that future robust prospective studies examining the relationship between male BMI and assisted reproduction technique outcome are still needed. Those studies should use standardized methods for measuring male BMI rather than relying on self-reported data, adhere to the WHO definitions of BMI categories, account for the influence of important clinical variables such as female age, female BMI, IVF and ICSI cycle characteristics including the number of embryos transferred and embryo quality, and use the live birth rate per cycle as their primary outcome. Additionally, studies evaluating the effect of weight loss and reduction of male BMI on modifying IVF–ICSI outcome are needed so that fertility specialists can provide appropriate counselling for couples embarking on treatment.
Acknowledgement We thank Mr Bolaji Coker, Senior data manager and Statistician, Guy’s and St Thomas ‘s NHS Foundation trust, London, for his suggestions in revising this article.
A R T I C L E
I N F O
Article history: Received 7 June 2017 Received in revised form 20 December 2017 Accepted 2 January 2018
Declaration: The authors report no financial or commercial conflicts of interest.
Keywords: ART Assisted reproduction Male BMI Male obesity
REFERENCES
Agarwal, A., Mulgund, A., Hamada, A., Chayatte, M.R., 2015. A unique view on male infertility around the globe. Reprod. Biol. Endocrinol. 13, 37. Andersson, A.M., Jorgensen, N., Main, K.M., Toppari, J., Rajpert-De Meyts, E., Leffers, H., Juul, A., Jensen, T.K., Skakkebaek, N.E., 2008. Adverse trends in male reproductive health: we may have reached a crucial ‘tipping point’. Int. J. Androl. Anifandis, G., Dafopoulos, K., Messini, C.I., Polyzos, N., Messinis, I.E., 2013. The BMI of men and not sperm parameters effect on embryo quality and the IVF outcome. Andrology 1, 85–89. Bakos, H.W., Henshaw, R.C., Mitchell, M., Lane, M., 2011. Paternal body mass index is associated with decreased blastocyst development and reduced live birth rates following assisted reproductive technology. Fertil. Steril. 95, 1700–1704. Braga, D.P., Halpern, G., Figueira Rde, C., Setti, A.S., Ianconelli, A., Jr., Borges, E., Jr., 2012. Food Intake and social habits in male patients and its relationship to Intracytoplasmic sperm Injection outcomes. Fertil. Steril. 97, 53–59. Campbell, J.M., Lane, M., Owens, J.A., Bakos, H.W., 2015. Paternal Obesity negatively affects male fertility and assisted reproduction outcomes: a systematic review and meta-analysis. Reprod. Biomed. Online 4, 998–1005. Centre for Reviews and Dissemination, 2001. CRD Report Number 4, 2nd edition. Undertaking systematic reviews of research on effectiveness. CRD’s guidance for carrying out or commissioning reviews. NHS Centre for Reviews and Dissemination, University of York, York, UK. Colaci, D.S., Afeiche, M., Gaskins, A.J., Wright, D.L., Toth, T.L., Tanrikut, C., Hauser, R., Chavarro, J.E., 2012. Men’s body mass index in relation to embryo quality and clinical outcomes in couples undergoing in vitro fertilization. Fertil. Steril. 98, 1193–1199. DerSimonian, R., Laird, N., 1986. Meta-analysis in clinical trials. Control. Clin. Trials 7, 177–188. Egger, M., Davey Smith, G., Schneider, M., Minder, C., 1997. Bias in meta-analysis detected by a simple graphical test. Br. Med. J. 315, 629–634. Eisenberg, M.L., Kim, S., Chen, Z., Sundaram, R., Schisterman, E.F., Buck Louis, G.M., 2014. The relationship between male BMI and waist circumference on semen quality: data from the LIFE study. Hum. Reprod. 29, 193–200. Hakonsen, L.B., Thulstrup, A.M., Aggerholm, A.S., Olsen, J., Bonde, J.P., Andersen, C.Y., Bungum, M., Ernst, E.H., Hansen, M.L., Ernst, E.H., Ramlau-Hansen, C.H., 2011. Does weight loss improve semen quality and reproductive hormones? Results from a cohort of severely obese men. Reprod. Health 8, 24. Higgins, J.P.T., Thompson, S.G., 2002. Quantifying heterogeneity in a meta-analysis. Stat. Med. 21, 1539–1558. Kahn, B.E., Brannigan, R.E., 2017. Obesity and male infertility. Curr. Opin. Urol. 27, 441–445. Katib, A., 2016. Mechanisms linking obesity to male infertility. Cent. European J. Urol. 68, 79–85. Keltz, J., Zapantis, A., Jindal, S.K., Lieman, H.J., Santoro, N., Polotsky, A.J., 2010. Overweight Men: clinical pregnancy after ART is decreased in IVF but not in ICSI cycles. J. Assist. Reprod. Genet. 27, 539–544.
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002
ARTICLE IN PRESS REPRODUCTIVE BIOMEDICINE ONLINE ■■ (2018) ■■–■■
Kupka, M.S., Gnoth, C., Buhler, K., Dahncke, W., Kruessel, J.S., 2010. Effect of female and male obesity on IVF–ICSI: results of 700,00 ART-cycles in Germany. Gynecol. Endocrinol. 0, 1–6. La Vignera, S., Condorelli, R.A., Vicari, E., Calogero, A.E., 2012. Negative effect of increased body weight on sperm conventional and nonconventional flow cytometric sperm parameters. J. Androl. 33, 53–58. Le, W., Su, S.H., Shi, L.H., Zhang, J.F., Wu, D.L., 2016. Effect of male body mass index on clinical outcomes following assisted reproductive technology: a meta-analysis in couples undergoing intracytoplasmic sperm injection. Andrologia 48, 406–424. Lewis, S., Clarke, M., 2001. Forest plots: trying to see the wood and the trees. BMJ 322, 1479–1480. Link, M.H., Schliep, K.C., Deshmukh, R., Porucznik, C.A., Carrell, D., Hammoud, A.O., 2012. Effect of paternal obesity on in vitro fertilization (IVF) pregnancy and live birth rates. Fertnstert 077, 746. Mantel, N., Haenszel, W., 1959. Statistical aspects of the analysis of data from retrospective studies of disease. J. Natl. Cancer Inst. 22, 719– 748. Merhi, Z.O., Keltz, J., Zapantis, A., Younger, J., Berger, D., Lieman, H.J., Jindal, S.K., Polotsky, A.J., 2013. Male adiposity impairs clinical pregnancy rate by In Vitro Fertilisation without affecting Day 3 embryo quality. Obesity (Silver Spring) 21, 1608–1612. Mushtaq, R., El-Toukhy, T., Khalaf, Y., Gebhardt, R., Rittenberg, V. 2016. Effect of male BMI on IVF outcome after elective single blastocyst transfer. A prospective study. Poster presented at European Society of Human Reproduction and Endocrinology annual meeting; July 2016 3–6.Helsinki, Finland. National Institute for Health and Care Excellence, 2013. Fertility Problems: Assessment and Treatment. NICE, London. Nguyen, D.M., El-Serag, H.B., 2010. The epidemiology of obesity. Gastroenterology 39, 1–7. Petersen, G.L., Schmidt, L., Pinborg, A., Kamper-Jorgensen, M., 2013. The influence of female and male body mass index on live births after assisted reproductive technology treatment: a nationwide register-based cohort study. Fertil. Steril. 99, 1654–1662. Popkin, B.M., Adair, L.S., Ng, S.W., 2012. Global nutrition transition and pandemic of obesity in developing countries. Nutr. Rev. 70, 3–21. Rittenberg, V., Seshadri, S., Sunkara, S.K., Sobaleva, S., Oteng-Ntim, E., El-Toukhy, T., 2011. Effect of body mass index on IVF treatment outcome: an updated systematic review and meta-analysis. Reprod. Biomed. Online 23, 421–439. Schliep, K.C., Mumford, S.L., Ahrens, K.A., Hotaling, J.M., Carrell, D.T., Link, M., Hinkle, S.N., Kissell, K., Porucznik, C.A.,
13
Hammoud, A.O., 2015. Effect of male and female body mass index on pregnancy and live birth success after in vitro fertilization. Fertil. Steril. 103, 388–395. Stroup, D., Berlin, J., Morton, S., Olkin, I., Williamson, G.D., Rennie, D., Moher, D., Becker, B.J., Sipe, T.A., Thacker, S.B., 2000. Metaanalysis of observational studies in epidemiology (MOOSE) group. J. Am. Med. Assoc. 283, 2008–2012. Sundaram, R., Sunni, L., Germaine, M., Buck, L., 2017. Couples’ body composition and time-to-pregnancy. Hum. Reprod. 32, 662–668. Thomsen, L., Humaidan, P., Bungum, L., Bungum, M., 2014. The effect of male overweight on semen quality and outcome of assisted reproduction. Asian J. Androl. 16, 749–754. Thonneau, P., Marchand, S., Tallec, A., Ferial, M.L., Ducot, B., Lansac, J., Lopes, P., Tabaste, J.M., Spira, A., 1991. Incidence and main causes of infertility in a resident population (1,850,000) of three French regions (1988–1989). Hum. Reprod. 6, 811–816. Tsao, C.W., Liu, C.Y., Chou, Y.C., Cha, T.L., Chen, S.C., Hsu, C.Y., 2015. Exploration of the association between obesity and semen quality in a 7630 male population. PLoS ONE 10. Umul, M., Kose, S.A., Bilen, E., Altuncu, A.G., Oksay, T., Guney, M., 2015. Effect of increasing paternal body mass index on pregnancy and live birth rates in couples undergoing intracytoplasmic sperm injection. Andrologia 47, 360–364. Wang, X., Hao, J., Zhang, F., Li, J., Kong, H., Guo, Y., 2016. Effect of female and male body mass indices on the treatment outcomes and neonatal birthweights associated with in vitro fertilization/ intracytoplasmic sperm injection treatment in China. Fertil. Steril. 106, 2–460. Wells, G., Shea, B., O’Connell, D. 2000. The Newcastle-Ottawa Scale (NOS) for assessing the quality of non-randomized studies in meta-analysis. Proceedings of the Third Symposium on Systematic Reviews. Beyond the Basics: improving Quality and Impact. 3—5 July, Oxford. WHO, 2017. Obesity and Overweight. World Health Organization, Geneva. http://www.who.int/mediacentre/factsheets/fs311/en/. Fact Sheet 311. Wu, Z., Lu, X., Wang, M., Cheng, H., 2015. Correlation between body mass index of chinese males and assisted reproductive technology outcome. Int. J. Clin. Exp. Med. 8, 21472. Zhao, J., Zhang, Q., Wang, Y., Li, Y., 2014. Whether sperm deoxyribonucleic acid fragmentation has an effect on pregnancy and miscarriage after in vitro fertilization/intracytoplasmic sperm injection: a systematic review and meta-analysis. Reprod. Biomed. Online 4, 998–1005.
Please cite this article in press as: Rabia Mushtaq, et al., Effect of male body mass index on assisted reproduction treatment outcome: an updated systematic review and meta-analysis, Reproductive BioMedicine Online (2018), doi: 10.1016/j.rbmo.2018.01.002