Background
Long-chain omega-3 fatty acids and their food sources have garnered interest as a potential nutrient with wide-range health benefits, including fertility.
Objective
This study aimed to investigate the association of women’s and men’s intake of omega-3 fatty acids and omega-3 rich–foods with semen quality and outcomes of infertility treatment with assisted reproductive technologies.
Study Design
Couples presenting to the Massachusetts General Hospital were invited to enroll in a prospective cohort study (2007–2020). Male and female diets were assessed using a validated 131-item food frequency questionnaire. The primary outcomes were implantation, clinical pregnancy, and live birth probabilities. The secondary outcomes included total and clinical pregnancy loss and conventional semen parameters, for males only. We estimated the relationship between intakes of omega-3 fatty acids, nuts, and fish and the probability (95% confidence interval) of study outcomes using generalized linear mixed models to account for repeated treatment cycles per participant while simultaneously adjusting for age, body mass index, smoking status, education, dietary patterns, total energy intake, and male partner diet.
Results
A total of 229 couples and 410 assisted reproductive technology cycles were analyzed for primary and secondary outcomes. Of note, 343 men contributing 896 semen samples were included in analyses for semen quality measures. Women’s docosahexaenoic acid + eicosapentaenoic acid intake was positively associated with live birth. The multivariable-adjusted probabilities of live birth for women in the bottom and top quartiles of eicosapentaenoic acid + docosahexaenoic acid intake were 0.36 (95% confidence interval, 0.26–0.48) and 0.54 (95% confidence interval, 0.42–0.66) ( P trend=.02). Eicosapentaenoic acid + docosahexaenoic acid intake was inversely related to the risk of pregnancy loss, which was 0.53 among women in the lowest quartile of eicosapentaenoic acid + docosahexaenoic acid intake and 0.05 among women in the highest quartile ( P trend=.01). Men’s intake of total omega-3 fatty acids was positively related to sperm count, concentration, and motility, but unrelated to any assisted reproductive technology outcomes. Similar associations were observed when evaluating the intake of primary food sources of these fatty acids.
Conclusion
Women’s consumption of omega-3 fatty acids and omega-3–rich foods may improve the probability of conception by decreasing the risk of pregnancy loss. In addition, men’s intake of omega-3 fatty acids may influence semen quality.
Introduction
Omega-3 fatty acids and their food sources, such as nuts and fish, have garnered interest for their potential influence on fertility. Previous studies have suggested that women’s intake of omega-3 fatty acids may be beneficial for fertility among couples without a history of infertility. In addition, we have reported that women’s serum omega-3 fatty acids and fish consumption are positively associated with live births during infertility treatment with assisted reproductive technologies (ARTs). Furthermore, previous work has suggested that men’s dietary omega-3 and fish consumption or fish oil supplement use is positively associated with semen quality. Moreover, data from randomized clinical trials (RCTs) show positive effects of nuts consumption and semen quality among men in the general population. , However, we have previously reported no association between men’s fish intake and outcomes of infertility treatment with ARTs.
Why was this study conducted?
Despite evidence suggesting that men’s and women’s intake of omega-3 fatty acids and omega-3–rich foods (nuts and fish) may have a positive influence on fertility, studies simultaneously considering the intakes of both prospective parents on fertility are limited.
Key findings
We found that women’s consumption of omega-3 fatty acids and omega-3–rich foods may improve the probability of live birth by decreasing the risk of pregnancy loss. In addition, men’s intake of omega-3 fatty acids may influence semen quality.
What does this add to what is known?
This study examined the association of men’s and women’s intake of omega-3 fatty acids and their primary food sources with couples’ assisted reproductive technology outcomes and men’s semen quality parameters in the same cohort of participants.
Despite evidence suggesting that men’s and women’s intake of omega-3 fatty acids and omega-3–rich foods may positively influence the couples’ fertility, studies simultaneously considering intakes of both prospective parents are limited. This raises questions about the extent to which associations with women’s intake may partly reflect associations with their partner’s intake given the degree of concordance in intake of food sources of omega-3 fatty acids in couples. This question has been addressed by 1 previous study among pregnancy planners, which found that the intake of seafood of both male and female partners was related to a shorter time to pregnancy. No previous study has addressed this issue in couples undergoing infertility treatment. Thus, this study aimed to investigate the association of women’s and men’s intake of omega-3 fatty acids and omega-3–rich foods (nuts and fish) with ART outcomes among couples undergoing assisted reproduction and men’s intake with semen quality.
Materials and Methods
Study population
Between 2004 and 2020, couples presenting to the Massachusetts General Hospital (MGH) Fertility Center for evaluation and treatment of infertility were invited to participate in the Environment and Reproductive Health (EARTH) Study, a prospective cohort study aimed at identifying environmental and nutritional determinants of fertility. Couples were eligible to participate if they were within the target age range (18–55 years for men and 18–45 years for women) and were planning to use their gametes for infertility treatment. Diet assessments were introduced in 2007. The study was approved by the institutional review boards of MGH and the Harvard T.H. Chan School of Public Health. All participants provided informed consent for participation.
For this analysis, we selected all couples where both partners completed a diet assessment before treatment and the female partner completed at least 1 ART cycle. Of the 462 couples who joined the study in 2007, 104 were excluded because the male partner did not complete a diet assessment. For evaluation of semen quality, 15 men were excluded because of azoospermia, leaving 896 semen samples from 343 men available for analysis. For ART outcomes evaluation, we excluded 113 couples because they were treated with intrauterine insemination and 16 couples whose treatment had started before diet assessment. The remaining 229 couples, who underwent 410 ART cycles, were included in the analysis ( Supplemental Figure 1 ). In these couples, 9 women had incomplete diet data, which were imputed using the median values in the overall study population to maximize statistical power.
Diet assessment
Diet was assessed using a validated 131-item food frequency questionnaire (FFQ). Participants were asked to report how often they consumed the listed foods and beverages during the previous year. The nutrient content of each item evaluated was calculated with the nutrient database of the US Department of Agriculture. We considered intake of (1) long-chain omega-3 (LCN-3) (defined as the sum of eicosapentaenoic acid [EPA], docosahexaenoic acid [DHA], total omega-3 fatty acids [defined as the sum of LCN-3], and alpha-linolenic acid [ALA]) and (2) foods rich in omega-3 fatty acids (nuts and fish). Total nut consumption was defined as the sum of peanuts, walnuts, and other nuts (eg, almonds and hazelnuts). Total fish consumption was defined as the sum of dark fish, white fish, and shellfish. In validation studies, the correlation between intake of omega-3 fatty acids estimated by FFQ and plasma levels spaced approximately 6 months apart was 0.62 (95% confidence interval [CI], 0.56–0.68), which is superior to the correlations comparing plasma levels with multiple 24-hour recalls and comparable to using 14 days of weighted diet records. The Prudent and Western dietary patterns were identified using principal component analysis as described elsewhere. Higher scores indicated higher adherence to the respective pattern.
Clinical management and outcomes assessment
At enrollment, weight and height were measured. Participants completed a detailed questionnaire with information on family, medical and reproductive, and occupational history and lifestyle factors.
Women underwent 1 of 3 stimulation protocols as clinically indicated: antagonist, flare, or luteal phase agonist. The primary outcomes of this study were implantation, clinical pregnancy, and live birth. Clinicians evaluated fertilization status on day 1 after fertilization based on the presence of 2 pronuclei. Embryo transfer was performed after stimulation and retrieval or after thawing of cryopreserved embryos. Implantation was assessed 14 to 15 days after embryo transfer and defined as a serum beta-human chorionic gonadotropin (β-hCG) level of >6 mIU/mL. Clinical pregnancy was defined as the presence of an intrauterine gestational sac confirmed by ultrasound at approximately 6 weeks of gestation. Live birth was defined as the birth of a neonate at or after 24 weeks of gestation. The secondary outcomes included total pregnancy loss, defined as a positive β-hCG that did not result in live births, and clinical pregnancy, defined as a clinical pregnancy that did not result in live births. For men, the secondary outcomes also included conventional semen parameters: ejaculate volume, sperm count and concentration, total and progressive motility, and morphology. Semen was collected by masturbation following a recommended 48-hour period of sexual abstinence. Some men provided multiple semen samples. Semen parameters were assessed on the basis of the 2010 World Health Organization manual guideline after 30 minutes of liquefaction using the computer-assisted semen analysis system (HTM-IVOS; Hamilton Thorne Ltd, Beverly, MA).
Statistical analyses
Participants were categorized into quartiles of total omega-3, total nuts, and total fish intake. Baseline participant characteristics were presented as median (interquartile range [IQR]) or number (percentage), and differences were compared across quartiles of total omega-3 intake using the Kruskal-Wallis test for continuous variables and the chi-squared or Fisher exact test for categorical variables. Spearman correlations were used to describe within-couple similarities in intakes of omega-3 fatty acids, nuts, and fish. For primary study outcomes (implantation, clinical pregnancy, and live birth), all women who started a cycle were included in the analysis. For secondary outcomes, total pregnancy loss and clinical pregnancy loss, only for women who achieved either a biochemical or a clinical pregnancy, were included in the analysis. We estimated the probability and 95% CI for ART outcomes by fitting multivariable generalized linear mixed models (GLMMs) with binomial (implantation, clinical pregnancy, live birth, and total and clinical pregnancy loss) distribution and random intercepts to account for repeated cycles. We estimated the marginal means (95% CI) for semen parameters by fitting multivariable GLMMs with repeated intercepts to account for repeated semen samples. Tests for linear trend were performed by modeling intake as a continuous variable where each man or woman was assigned the median intake of each category.
Our primary analysis focused on absolute intake using the standard multivariable method to adjust for total energy intake. To evaluate the robustness of our findings, we performed sensitivity analyses using the nutrient residual method and the multivariable energy density method to adjust for total energy intake. Confounding factors were evaluated using previous knowledge on biologic relevance and descriptive statistics from our cohort ( Table 1 ). Fully adjusted models for men’s omega-3 associations included terms for male age (years), body mass index (BMI; kg/m 2 ), energy intake (kcal/day), education (high school or less, college or higher), Prudent and Western patterns, and female omega-3 intake. Fully adjusted models for women’s omega-3 associations included terms for female age, BMI, energy intake, smoking status (never or ever), education, Prudent and Western patterns, and male omega-3 intake. Fully adjusted models for men’s nuts and fish associations included terms for male age and BMI, male energy intake, male education, male Prudent and Western patterns, female total fish intake (for nuts models), or female total nuts intake (for fish models). Fully adjusted models for women’s nuts and fish associations included terms for female age and BMI, female energy intake, female smoking status, female education, female Prudent and Western patterns, male total fish intake (for nuts models), or male total nuts intake (for fish models). The final adjusted multivariable models for semen quality parameters included age and BMI, energy intake, physical activity (minutes/week), race (White or other), smoking status, and sexual abstinence time (days). Of note, we adjusted for physical activity in models for semen quality but not in models for ART outcomes because we have previously reported that men’s physical activity is related to semen quality in this and other cohorts, but men’s and women’s physical activity are unrelated to ART outcomes in this cohort. , To test the robustness of our findings to missing data assumptions of GLMMs and imputation of missing diet data, we repeated our analyses using cluster-weighed generalized estimating equation (CW-GEE) models and excluding women with incomplete diet data. Statistical analyses were performed using SAS (version 9.4; SAS Institute, Cary, NC).
| Characteristic | Overall | Men’s total omega-3 intake (g/d) | P value | Women’s total omega-3 intake (g/d) | P value | ||
|---|---|---|---|---|---|---|---|
| Q1 (1.06–3.12) | Q4 (4.77–9.49) | Q1 (1.27–2.64) | Q4 (4.45–8.23) | ||||
| n | 229 | 57 | 57 | 57 | 57 | ||
| Male demographic characteristics | |||||||
| Age (y) | 36.0 (33.6–39.3) | 36.0 (33.6–38.8) | 37.2 (34.4–40.5) | .39 | 36.0 (32.9–38.5) | 36.6 (33.7–40.4) | .40 |
| BMI (kg/m 2 ) | 27.0 (24.3–28.9) | 27.8 (25.4–29.7) | 27.8 (25.7–29.5) | .03 | 27.0 (24.3–29.1) | 27.0 (24.1–29.3) | .89 |
| Race, White | 211 (92) | 51 (22) | 54 (24) | .76 | 51 (22) | 53 (23) | .82 |
| Smoking status, ever smoker | 44 (19) | 12 (5) | 13 (6) | .76 | 7 (3) | 12 (5) | .43 |
| Education, college or higher | 199 (87) | 48 (21) | 47 (21) | .34 | 52 (23) | 49 (21) | .73 |
| Moderate-to-vigorous physical activity (min/wk) | 180.0 (60.0–390.0) | 161.5 (44.0–329.5) | 296.0 (118.5–390.0) | .28 | 150.0 (41.5–341.0) | 239.0 (112.5–390.0) | .34 |
| Sexual abstinence (d) | 2.4 (2.4–2.5) | 2.4 (1.8–3.4) | 2.5 (2.0–3.4) | .77 | 2.4 (2.4–2.5) | 2.4 (2.4–2.6) | .55 |
| Male dietary parameters | |||||||
| Energy intake (kcal/d) | 1937 (1592–2414) | 1402 (1152–1664) | 2623 (2165–2991) | <.001 a | 1741 (1468–2191) | 2144 (1731–2534) | <.001 a |
| Prudent pattern score | −0.1 (−0.6 to 0.6) | −0.6 (−1.0 to −0.3) | 0.7 (−0.1 to 1.8) | <.001 a | −0.5 (−0.8 to 0.5) | 0.0 (−0.5 to 0.9) | .04 a |
| Western pattern score | −0.2 (−0.7 to 0.5) | −0.7 (−1.1 to −0.4) | 0.7 (−0.3 to 1.2) | <.001 a | −0.6 (−1.1 to 0.0) | 0.1 (−0.5 to 1.0) | <.001 a |
| Total omega-3 | 3.8 (3.2–4.8) | 2.7 (2.2–2.9) | 5.6 (5.2–6.8) | <.001 a | 2.9 (2.3–3.7) | 4.0 (3.5–4.8) | <.001 a |
| Female demographic characteristics | |||||||
| Age (y) | 35.0 (32.0–38.0) | 35.0 (31.0–38.0) | 35.0 (33.0–38.0) | .74 | 35.0 (32.0–37.0) | 35.0 (32.0–37.0) | .28 |
| BMI (kg/m 2 ) | 23.0 (21.0–25.7) | 23.5 (20.8–25.7) | 22.7 (20.9–25.0) | .39 | 22.6 (20.8–24.2) | 23.8 (21.6–27.9) | .28 |
| Race, White | 194 (85) | 45 (20) | 47 (21) | .37 | 46 (20) | 49 (21) | .81 |
| Smoking status, ever smoker | 59 (26) | 11 (5) | 19 (8) | .15 | 12 (5) | 14 (6) | .57 |
| Education, college or higher | 214 (94) | 54 (24) | 53 (23) | .84 | 52 (23) | 55 (24) | .47 |
| Moderate-to-vigorous physical activity (min/wk) | 149.0 (29.5–298.5) | 114.0 (12.0–210.0) | 150.0 (29.5–332.5) | .03 a | 113.5 (12.0–210.0) | 191.5 (60.0–392.5) | .02 a |
| Female dietary parameters | |||||||
| Energy intake (kcal/d) | 1675 (1319–2004) | 1492 (1148–1694) | 1796 (1683–2142) | <.001 a | 1148 (994–1369) | 2158 (1841–2514) | <.001 a |
| Prudent pattern score | −0.3 (−0.7 to 0.3) | −0.4 (−0.8 to 0.2) | 0.0 (−0.5 to 0.4) | .02 a | −0.7 (−1.0 to −0.4) | 0.4 (−0.1 to 0.8) | <.001 a |
| Western pattern score | −0.2 (−0.7 to 0.5) | −0.4 (−0.9 to 0.0) | 0.2 (−0.4 to 1.0) | <.001 a | −0.8 (−1.0 to −0.6) | 0.3 (−0.2 to 1.1) | <.001 a |
| Total omega-3 | 3.5 (2.7–4.4) | 3.5 (2.7–3.8) | 4.5 (3.8–5.9) | <.001 a | 2.1 (1.7–2.4) | 5.0 (4.8–5.8) | <.001 a |
| Baseline cycle characteristics | |||||||
| Infertility diagnosis | .29 | .91 | |||||
| Female factor | 79 (35) | 21 (9) | 17 (7) | 18 (8) | 18 (8) | ||
| Male factor | 87 (38) | 25 (11) | 18 (8) | 23 (10) | 21 (9) | ||
| Unexplained | 63 (28) | 11 (5) | 22 (10) | 16 (7) | 18 (8) | ||
| Treatment protocol | .62 | .96 | |||||
| Antagonist | 32 (14) | 10 (4) | 7 (3) | 10 (4) | 8 (4) | ||
| Flare | 19 (8) | 7 (3) | 4 (2) | 5 (2) | 4 (2) | ||
| Luteal phase agonist | 160 (70) | 37 (16) | 41 (18) | 37 (16) | 41 (18) | ||
| Egg donor or cryocycle | 18 (8) | 3 (1) | 5 (2) | 5 (2) | 4 (2) | ||
| Embryo transfer day | .22 | .21 | |||||
| Day 2 | 10 (5) | 2 (1) | 5 (3) | 3 (2) | 1 (1) | ||
| Day 3 | 80 (42) | 22 (12) | 19 (10) | 12 (6) | 28 (15) | ||
| Day 5 | 101 (53) | 26 (14) | 23 (12) | 29 (15) | 22 (12) | ||
| Number of embryos transferred | .40 | .75 | |||||
| 1 embryo | 66 (35) | 19 (10) | 11 (6) | 16 (8) | 19 (10) | ||
| 2 embryos | 97 (51) | 23 (12) | 31 (16) | 23 (12) | 22 (12) | ||
| ≥3 embryos | 27 (14) | 7 (4) | 5 (3) | 5 (3) | 9 (5) | ||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
