Maternal whole grain intake and outcomes of in vitro fertilization

Abstract

Objective: To evaluate the relationship between pretreatment intake of whole grains and outcomes of IVF.

Design: Prospective cohort study.

Setting: Academic medical center.

Patient(s): A total of 273 women who collectively underwent 438 IVF cycles.

Intervention(s): Whole grain intake was assessed with a validated food frequency questionnaire at enrollment.

Main outcome measure(s): Intermediate and clinical end points of IVF were abstracted from medical records.

Result(s): Women had a median whole grain intake of 34.2 g per day (∼1.2 servings/day). Higher pretreatment whole grain intake was associated with higher probability of implantation and live birth. The adjusted percentage of cycles resulting in live birth for women in the highest quartile of whole grain intake (>52.4 g/day) was 53% (95% confidence interval [CI] 41%, 65%) compared with 35% (95% CI 25%, 46%) for women in the lowest quartile (<21.4 g/day). This association was largely driven by intake of bran as opposed to germ. When intermediate end points of IVF were examined, only endometrial thickness on the day of ET was associated with whole grain intake. A 28-g per day (∼1 serving/day) increase in whole grain intake was associated with a 0.4-mm (95% CI 0.1, 0.7 mm) increase in endometrial thickness.

Conclusion(s): Higher pretreatment whole grain intake was related to higher probability of live birth among women undergoing IVF. The higher probability of live birth may result from increased endometrial thickness on the day of ET and improved embryo receptivity manifested in a higher probability of implantation.

Keywords: Whole grains; assisted reproductive technology; female infertility; in vitro fertilization.

Figure 1

Association between whole grain intake and in vitro fertilization (IVF) outcomes per initiated cycle in 273 women (438 IVF cycles) in the Environment and Reproductive Health (EARTH) Study. All analyses were conducted using generalized linear mixed models with random intercepts, binomial distribution, and logit link function. Data are presented as marginal mean probabilities adjusted for calorie intake, age, BMI, race, dietary patterns, and folate and alcohol intake. *indicates that the p-value for the comparison vs. Q1 was <0.05.

Figure 2

Predicted mean endometrial thickness with increasing whole grain intake in 255 women (352 fresh IVF cycles with egg retrieval) from the Environment and Reproductive Health Study. All analyses were conducted using mixed models with random intercepts, normal distribution, and identity link function. Data are presented as marginal means adjusted for calorie intake, age, BMI, race, dietary patterns, and folate and alcohol intake.