Introduction
Infertility, defined as the inability to achieve pregnancy after 12 months or more of regular unprotected intercourse, can result from male, female, or unexplained factors. Unexplained infertility accounts for approximately 30% of cases [1,2]. In addition to being a medical condition, infertility has substantial sociological, psychological, and economic consequences, affecting marital relationships, quality of life, and mental well-being [3].
Female fertility declines notably after age 30, primarily because of reduced ovarian reserve and declining oocyte quality [4]. The evaluation and treatment of infertility require a multidisciplinary approach that considers both male and female factors. In vitro fertilization (IVF) success is influenced by multiple factors, including maternal age, the cause and duration of infertility, ovarian reserve, uterine receptivity, oocyte and embryo quality, and semen parameters. Environmental, emotional, and lifestyle factors also contribute to reproductive outcomes.
Among lifestyle-related factors, smoking is widely recognized as a major risk factor for infertility. The prevalence of smoking has increased among both men and women of reproductive age, and tobacco exposure has been associated with adverse effects on oocyte quality, sperm parameters, fertilization rates, and IVF success [5,6]. Cigarette smoke contains mutagenic and carcinogenic compounds, including cotinine and benzo[a]pyrene, which may impair ovarian folliculogenesis, steroidogenesis, embryo implantation, and uterine vascularization [7,8]. In women, smoking disrupts follicular development and has been associated with oxidative stress, increased apoptosis, impaired oocyte–granulosa cell interactions, and loss of nuclear function [9].
Notably, the region in which this study was conducted has a relatively high prevalence of smoking among individuals of reproductive age. National and regional epidemiological data indicate that tobacco exposure rates in Central Anatolia, Türkiye, exceed national averages, particularly among men, while smoking rates among women have also been increasing. This high background prevalence underscores the importance of assessing the reproductive effects of smoking and antioxidant use within this population.
Another key factor in fertility is oxidative stress, which has been implicated in 30% to 80% of infertility cases [10,11]. Excessive reactive oxygen species (ROS) can damage sperm DNA, reduce oocyte quality, and impair embryo implantation [12]. Antioxidants obtained through diet or supplementation have been proposed as a strategy to reduce oxidative stress and improve gamete quality [13]. Despite extensive research examining smoking and antioxidant use separately, their combined effects on IVF outcomes remain unclear. The present study is notable in that it evaluates both partners simultaneously, investigating how smoking and antioxidant use may influence IVF success and live birth rates beyond the organic causes of infertility.
Methods
1. Research design
This retrospective cohort study included couples who underwent IVF treatment between January 2018 and January 2019 at a tertiary infertility center. A quantitative retrospective cohort design was used. Only heterosexual couples undergoing IVF were included; therefore, gender and sex variables were not differentiated.
2. Study population and setting
This retrospective cohort study included 327 couples who underwent IVF treatment at a tertiary infertility center between 2018 and 2019. The study population consisted of women aged 19 to 45 years and their partners, all of whom had documented records of smoking status and antioxidant use before and during treatment. Patients with additional pathologies that could affect fertility treatment outcomes were excluded.
Initially, 372 couples were enrolled in the study; however, 45 couples were excluded due to incomplete treatment data. The final analysis thus included 327 couples who met the eligibility criteria. Exclusion criteria included patients younger than 19 years or older than 45 years and those lacking recorded data on smoking status or antioxidant use.
3. Group classification
Participants were classified into four categories according to smoking status, antioxidant use before treatment, and antioxidant use during treatment: none, male only, female only, and both partners. This couple-based classification was applied consistently throughout all analyses.
4. Variables and measurements
The study variables included age, weight, height, smoking status and duration, antioxidant use, the cause and duration of infertility, basal hormone levels—including follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), thyroid-stimulating hormone (TSH), prolactin, anti-Müllerian hormone (AMH), and progesterone—as well as antral follicle count (AFC), semen concentration and motility, stimulation protocol, total oocyte count, number of metaphase II oocytes, embryo number and quality, number of embryos transferred, clinical pregnancy, and live birth.
5. Exposure definitions: smoking and antioxidant supplementation
Smoking status was determined based on self-reported cigarette use by either partner and was categorized as none, male only, female only, or both partners. When available, the duration of smoking (years) and the average daily number of cigarettes smoked were recorded. Because biochemical verification (e.g., cotinine measurement) was not feasible in this retrospective study, potential underreporting should be considered a limitation.
Antioxidant and multivitamin supplementation was defined as any self-reported use of over-the-counter preparations within the 3 months preceding IVF treatment (‘pre-treatment’) or during controlled ovarian stimulation (‘during treatment’). These supplements were primarily fertility support formulations used in routine clinical practice and typically contained combinations of L-carnitine, L-arginine, coenzyme Q10, myo-inositol and/or D-chiro-inositol, folic acid, omega-3 fatty acids, vitamins C, E, and B-complex (B1–B12), and trace elements such as zinc, selenium, and magnesium. In a minority of women, dehydroepiandrosterone preparations were also used as an adjunct to improve ovarian response.
Although the approximate duration of supplement use (in months) was recorded for many patients, detailed information regarding the exact dose and brand-specific formulation of antioxidant preparations was heterogeneous and not standardized. Because these supplements are not reimbursed and must be purchased out of pocket, the choice of product, adherence, and duration of use likely varied according to individual financial circumstances. Due to this variability, antioxidant supplementation was analyzed as a categorical exposure (none, male only, female only, or both partners) rather than according to specific dose or duration. This heterogeneity represents a key limitation and may have influenced the interpretation of antioxidant-related associations.
6. Statistical analysis
Statistical analyses were conducted using SPSS ver. 21.0 (IBM Corp.). Categorical variables were presented as numbers and percentages, whereas continuous variables were summarized as mean±standard deviation or as median (minimum–maximum), depending on their distribution.
Normality was assessed using the Kolmogorov–Smirnov and Shapiro–Wilk tests. Comparisons between categorical variables were performed using the Pearson chi-square test. Continuous variables that did not follow a normal distribution were compared using the Mann–Whitney U test for two independent groups and the Kruskal–Wallis test for more than two independent groups. A p-value of less than 0.05 was considered to indicate statistical significance.
The sample size was estimated a priori using G*Power ver. 3.0 (University of Düsseldorf). Assuming an effect size of 0.3, a significance level (α) of 0.05, and 80% power, the minimum required sample size was calculated as 343 couples. The final cohort included 327 couples, which is slightly below the estimated requirement and may therefore limit statistical power.
The study protocol was approved by the No. 2 Clinical Research Ethics Committee of Ankara Bilkent City Hospital (Approval No.: E2-21-750; September 1, 2021). All procedures adhered to the principles of the Declaration of Helsinki. The requirement for written informed consent was waived by the Ethics Committee because of the retrospective design of the study.
Results
1. Participant demographics
The general characteristics and medical history of the 327 couples are summarized in Table 1. Overall, smoking was reported in approximately two-thirds of couples, most commonly among male partners. Antioxidant supplementation before treatment was more frequently reported by men than by women, whereas supplementation during treatment was reported predominantly by female partners.
2. Evaluation of ovarian reserve and basal hormone levels according to smoking status and antioxidant use
The associations between smoking status, antioxidant use, and ovarian reserve or basal hormonal parameters were evaluated. No significant differences were observed in serum AMH, FSH, LH, or TSH levels according to smoking status (p>0.05). However, E2 levels were significantly higher among smokers than among non-smokers (p=0.008).
When pre-treatment antioxidant use was examined, couples reporting antioxidant supplementation before IVF had significantly higher AMH levels than non-users, whereas other ovarian reserve markers and basal hormonal parameters were comparable between groups. Detailed results are provided in Supplementary Table 1.
In a subgroup analysis evaluating antioxidant supplementation initiated during IVF treatment, no significant differences were observed in AMH, AFC, FSH, LH, E2, prolactin, or TSH levels (p>0.05). Because no statistically significant findings were identified, these results are summarized in the text and are not presented in a separate table.
3. Clinical pregnancy and live birth outcomes according to smoking status
Clinical pregnancy and live birth outcomes according to couples’ smoking status are presented in Table 2. Clinical pregnancy occurred in approximately one-third of all couples. Pregnancy rates were slightly higher among non-smoking couples (34.3%) and couples in which only the male partner smoked (36.5%) than among couples in which both partners smoked (27.1%); however, these differences were not statistically significant.
Similarly, live birth rates were lowest among couples in which both partners smoked (12.9%) and highest among couples in which only the female partner smoked (30.8%), although no statistically significant association was observed between smoking status and live birth. Overall, smoking status was not significantly associated with clinical pregnancy or live birth.
4. Clinical pregnancy and live birth outcomes according to antioxidant use
Clinical pregnancy and live birth outcomes according to pre-treatment antioxidant use are presented in Table 2. Couples in which only the male partner used antioxidants before treatment had higher rates of pregnancy (45.5%) and live birth (35.1%) than the other groups. Couples in which only the female partner used antioxidants before treatment had the lowest pregnancy and live birth rates.
No significant differences in clinical pregnancy or live birth were observed among the remaining antioxidant use categories. Detailed IVF laboratory parameters according to pre-treatment antioxidant use are provided in Supplementary Table 2.
Antioxidant use initiated during IVF treatment was not associated with significant differences in either clinical pregnancy or live birth (Table 2). No meaningful variation in pregnancy or live birth outcomes was observed among couples according to antioxidant use during treatment. Detailed IVF laboratory parameters according to antioxidant use during treatment are presented in Supplementary Table 2.
5. Secondary analyses of semen parameters
Semen parameters did not differ significantly according to smoking status. Descriptive analyses of sperm motility and total motile sperm count according to antioxidant use before and during treatment are provided in Supplementary Table 3. To further characterize male smoking exposure, cumulative smoking intensity (pack-years) was calculated using daily cigarette consumption and smoking duration. The association between pack-years and semen parameters was assessed using Spearman correlation analysis. No statistically significant correlations were observed between pack-years and sperm motility (r=0.068, p=0.255) or total motile sperm count (r=0.035, p=0.555).
6. Multivariable logistic regression analysis
In the univariate analyses, the number of embryos transferred and the total number of embryos formed were significantly associated with both clinical pregnancy and live birth. For clinical pregnancy, each additional embryo transferred was associated with higher odds of pregnancy (odds ratio [OR], 2.708; 95% confidence interval [CI], 1.797 to 4.080; p<0.001), and the total number of embryos formed was also positively associated with clinical pregnancy (OR, 1.256; 95% CI, 1.148 to 1.374; p<0.001) (Table 3). Similarly, for live birth, the total number of embryos formed remained significant in the univariate analysis (OR, 1.226; 95% CI, 1.119 to 1.343; p<0.001) (Table 4).
In the multivariable logistic regression analysis, after adjustment for potential confounders, the number of embryos transferred (adjusted odds ratio [aOR], 2.006; 95% CI, 1.272 to 3.166; p=0.003) and the total number of embryos formed (aOR, 1.190; 95% CI, 1.082 to 1.309; p<0.001) remained independent predictors of clinical pregnancy (Table 3). For live birth, the total number of embryos formed also remained a significant positive predictor (aOR, 1.192; 95% CI, 1.076 to 1.320; p<0.001) (Table 4). Increasing female age was associated with lower odds of live birth (aOR, 0.891; 95% CI, 0.812 to 0.979; p=0.016).
Pre-treatment antioxidant use by male partners displayed significance in the univariate analysis for clinical pregnancy (OR, 1.875; p=0.022) and live birth (OR, 2.200; p=0.008); however, these associations were not significant in the multivariable models. Smoking by both partners was negatively associated with live birth in the univariate analysis (OR, 0.426; p=0.043), and this association remained significant after adjustment (aOR, 0.374; 95% CI, 0.153 to 0.914; p=0.031).
Discussion
In this retrospective cohort study of 327 couples undergoing IVF, lifestyle-related exposures, including smoking status and antioxidant use, were evaluated in relation to clinical pregnancy and live birth. Overall, smoking status was not significantly associated with clinical pregnancy or live birth, although smoking by both partners was independently associated with reduced odds of live birth in the multivariable analysis. Pre-treatment antioxidant use, particularly when reported by male partners, was associated with pregnancy outcomes in the univariate analyses; however, this association did not persist after adjustment for relevant clinical and laboratory confounders. In contrast, laboratory-related parameters, including the number of embryos formed and the number of embryos transferred, were the strongest independent predictors of clinical pregnancy and live birth. These findings suggest that embryological factors may play a more central role in IVF success than the lifestyle-related exposures examined in this study.
Infertility imposes substantial psychological, social, and economic burdens on couples, and lifestyle-related factors, particularly smoking, have been widely studied in relation to assisted reproductive technology outcomes. Previous studies have shown that smoking may reduce ovarian reserve and increase the risk of implantation failure through the accumulation of tobacco-derived toxins in follicular fluid and the induction of oxidative stress [11,12]. However, in the present study, no significant differences were observed between smokers and non-smokers with respect to oocyte quality, embryo development, or pregnancy outcomes.
These findings may be explained by the dose-dependent effects of smoking, interindividual variability in susceptibility to tobacco exposure, and methodological limitations in exposure assessment. Although smoking has been associated with reduced AMH levels and impaired ovarian response in women undergoing IVF, as well as adverse effects on sperm DNA integrity, concentration, motility, and morphology in men [13-18], these biological effects may not consistently translate into clinical IVF outcomes.
In this cohort, smoking was more common among male partners. Higher serum E2 levels observed among couples in which both partners smoked may reflect a hormonal response to combined tobacco exposure. The reliance on self-reported smoking status, the limited number of female smokers, and the absence of quantitative or objective exposure measures such as serum or urinary cotinine levels likely contributed to these largely neutral findings, which differ from some previously reported associations in the literature. Given the established role of smoking as a major source of oxidative stress, the potential contribution of oxidative imbalance to reproductive outcomes warrants further investigation.
Oxidative stress has been implicated in several causes of female infertility, including endometriosis and polycystic ovary syndrome [18,19]. Experimental and clinical studies suggest that excessive ROS may impair oocyte competence, fertilization, and early embryonic development through alterations in gene expression and mitochondrial function [20-24].
Previous studies have demonstrated elevated ROS levels and reduced antioxidant capacity in subfertile men, particularly in the presence of conditions such as varicocele or increased scrotal temperature [25-28]. However, conflicting evidence remains regarding the central role of oxidative stress or DNA fragmentation in idiopathic male infertility [26,29,30]. Reflecting this uncertainty, a Cochrane meta-analysis concluded that although antioxidant therapy may improve pregnancy-related outcomes, the overall quality of the evidence remains low, underscoring the need for well-designed prospective studies [31].
This study has several limitations that should be acknowledged. Although 327 couples were categorized into four groups to examine the effects of smoking and antioxidant use on IVF outcomes, detailed subgroup stratification by infertility etiology was not performed, limiting the ability to assess antioxidant efficacy within specific clinical subpopulations. In addition, antioxidant users were not analyzed separately by smoking status, which complicates the interpretation of differences observed in laboratory parameters such as two-pronuclear zygote count, metaphase II oocyte count, and embryo number, particularly among couples in which only the male partner reported antioxidant use.
Smoking status and antioxidant use were assessed retrospectively based on participant self-report, introducing potential recall bias and exposure misclassification, particularly among light or intermittent smokers. Objective biochemical measures of smoking exposure, such as cotinine testing, were not available. In addition, strict sperm morphology parameters assessed using the Kruger criteria were not routinely recorded, limiting comprehensive assessment of the biological effects of male smoking on semen quality. Antioxidant use by male partners may also reflect poorer baseline semen quality, raising the possibility of confounding by indication and limiting causal interpretation of the observed associations. Antioxidant supplementation was not standardized with respect to type, dosage, duration, or timing within the IVF cycle, contributing to heterogeneity that limits interpretability and reproducibility.
In conclusion, this study provides a comprehensive evaluation of lifestyle-related factors by simultaneously assessing smoking and antioxidant use in both male and female partners undergoing IVF. In adjusted analyses, embryological parameters, particularly the number of embryos formed and transferred, were the strongest independent predictors of clinical pregnancy and live birth, whereas smoking by both partners was associated with reduced live birth rates.
Although pre-treatment antioxidant use, especially when reported by male partners, showed favorable associations in the univariate analyses, this effect did not persist after multivariable adjustment. Therefore, these findings should be interpreted as hypothesis-generating rather than as evidence of an independent clinical benefit.
Given the retrospective design, the heterogeneity of antioxidant formulations, and the reliance on self-reported exposures, causal inferences cannot be drawn. Well-designed prospective studies using standardized antioxidant protocols and objective biomarkers of oxidative stress are needed to clarify the roles of antioxidant supplementation and smoking-related oxidative imbalance in IVF outcomes.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement1
