Estimating the effects of hypothetical preconception dietary patterns interventions on in vitro fertilization and...

Among women undergoing in vitro fertilization/intracytoplasmic sperm injection (IVF/ICSI), the association between preconception dietary patterns and assisted reproductive outcomes remains unclear. We aimed to estimate the potential changes in assisted reproductive outcomes under hypothetical preconception dietary pattern interventions using the parametric g-formula.

In this prospective cohort study, 1414 infertile women aged 20 years or older were recruited using convenience sampling during pretreatment evaluation at a tertiary reproductive specialist hospital in Shandong Province between March and September 2024. Women's pretreatment diet was assessed using the simplified 25-item food frequency questionnaire. Six dietary patterns were calculated, including the energy-adjusted dietary inflammatory index (E-DII), Mediterranean diet (MD) score, American Heart Association (AHA) dietary index, Trichopoulou Mediterranean diet (TMD) score, alternate Mediterranean diet (AMED) score, and Dietary Approaches to Stop Hypertension (DASH) score. Outcomes included three categories: laboratory and clinical outcomes, pregnancy outcomes, and neonatal outcomes. Laboratory and clinical outcomes included implantation rate, blastocyst formation rate, good-quality blastocyst formation rate, and live birth rate. Pregnancy outcomes included miscarriage, gestational diabetes (GD), hypertensive disorders of pregnancy (HDPs), and preterm birth. Neonatal outcomes included large for gestational age (LGA), small for gestational age, and abnormal birth weight. Using the parametric g-formula, we simulated interventions increasing dietary scores to moderate adherence (75th percentile), high adherence (90th percentile), and optimal adherence, and estimated risk differences (RDs), risk ratios, and relative changes for these outcomes.

Before the simulated intervention, the rates were 17.35% for miscarriage, 34.01% for live birth, 14.10% for GD, 8.55% for HDP, and 23.31% for LGA. Under moderate adherence to the E-DII simulated intervention, the miscarriage risk was significantly lowered (11.41; 95% confidence interval [CI], 9.61-13.77), with an even lower risk under high adherence (8.66; 95% CI, 7.73-10.23) and the lowest risk under the optimal adherence strategy (7.11; 95% CI, 7.02-7.39). The corresponding RDs were -5.94 (95% CI, -8.43 to -3.34), -8.69 (95% CI, -11.34 to -6.05), and -10.24 (95% CI, -12.86 to -7.59), respectively. The E-DII simulated intervention showed consistent favorable patterns for live birth and GD. Under moderate adherence to the MD simulated intervention, the HDP risk was significantly lowered, with a lower risk under high adherence and the lowest risk under the optimal adherence strategy. The simulated HDP risks were 6.49 (95% CI, 2.24-10.53), 5.82 (95% CI, 0.64-10.48), and 4.94 (95% CI, 0.00-8.71), respectively. The corresponding RDs were -2.06 (95% CI, -4.57 to -0.00), -2.73 (95% CI, -6.17 to -0.05), and -3.61 (95% CI, -7.78 to -1.82), respectively. Under moderate adherence to the TMD simulated intervention, the LGA risk was significantly lowered, with a lower risk under high adherence and the lowest risk under the optimal adherence strategy. The simulated LGA risks were 22.12 (95% CI, 16.69-25.51), 21.28 (95% CI, 14.05-24.78), and 19.15 (95% CI, 7.08-23.10), respectively. The corresponding RDs were -1.19 (95% CI, -3.46 to -1.11), -2.03 (95% CI, -6.10 to -1.84), and -4.16 (95% CI, -13.07 to -3.52), respectively. No clear associations were observed for the AHA index, AMED score, or DASH score with the outcomes.

Simulated intervention analyses suggested that lower E-DII, reflecting reduced dietary inflammatory potential, was associated with select reproductive outcomes, particularly lower miscarriage risk and higher live birth under optimal adherence scenarios. Higher adherence to MD and TMD was also associated with lower simulated risks of HDP and LGA, respectively. These findings may help inform future research on preconception dietary optimization among women undergoing IVF/ICSI.