First-trimester protein intake linked to lower odds of small-for-gestational-age birth

A large US pregnancy cohort suggests that a middle range of first-trimester protein intake may be associated with lower odds of small-for-gestational-age birth, but inconsistent supporting analyses mean the finding is not yet ready to guide dietary recommendations.

Study: Association Between Maternal Protein Intake and the Risk of Adverse Pregnancy Outcomes. Image Credit: AYO Production / Shutterstock

A secondary analysis of data from a large prospective cohort of nulliparous women found that first-trimester protein intake of 0.8 to less than 1.1 g/kg/day was associated with lower adjusted odds of delivering a small-for-gestational-age (SGA) infant than intake of 1.1 g/kg/day or more. The paper was published in the journal Nutrients.

Background

An individual’s nutritional requirements change at different stages of the life course. During pregnancy, nutritional requirements increase to support both maternal health and fetal growth. Nutritional imbalances during this period are linked to adverse maternal and fetal outcomes.

Adequate supplies of macronutrients, especially protein, are essential for health and to support fetal growth and development. Protein is the source of amino acids that are key to all physiological processes, including growth, repair, protein synthesis, immune function, lipid and carbohydrate metabolism, and energy production.

Appropriate protein intake during pregnancy is linked to a reduced risk of adverse fetal growth outcomes and preterm birth. Long-term health issues have also been associated with both inadequate and excessive maternal protein intake during pregnancy. These are possibly linked to abnormalities in fetal developmental programming and epigenetic programming, leading to cardiometabolic ill-health in later life.

Public health recommendations have established various nutritional benchmarks, including estimated average requirements (EAR) and recommended dietary allowance (RDA). During pregnancy, the RDA for protein intake is 1.1 g/kg/day, higher than for non-pregnant women. The EAR is 0.88 g/kg/day.

There is little information about how protein intake in pregnancy is linked to pregnancy outcomes in the USA. The current study sought to bridge this gap.

Study characteristics

The data for this secondary analysis came from the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be (nuMoM2b), a prospective cohort study conducted at multiple sites. The study included 7,067 nulliparous participants with first-trimester dietary data whose pregnancies continued to at least 20 weeks and who had the required outcome data.

The researchers compared pregnancy outcomes across the four maternal protein intake quartiles. The pregnancy outcomes examined here included hypertensive disorders of pregnancy (HDP), gestational diabetes mellitus (GDM), large for gestational age (LGA), and small for gestational age (SGA).

Food intake was assessed once using the food frequency questionnaire between 6 and 13 weeks of gestation, 6 days prior to delivery. Using this, participants were divided into four protein intake quartiles: less than 0.6 g/kg/day, 0.6 to less than 0.8 g/kg/day, 0.8 to less than 1.1 g/kg/day, and 1.1 g/kg/day or higher.

The researchers adjusted the analyses for maternal age, body mass index (BMI), education, poverty level, insurance status, chronic hypertension, and daily energy intake.

Protein intake varied across demographic and socioeconomic groups

Most participants were non-Hispanic White and aged 18 to 34 years. Most were employed and had at least a bachelor’s degree, were covered by commercial insurance, and had income levels above the federal poverty line.

White participants were somewhat more represented in the second quartile, 0.6 to less than 0.8 g/kg per day. Hispanic/Latina and Asian participants were more highly represented in the highest intake category (at least 1.1 g/kg/day). Black and multiracial participants were more highly represented in the lowest intake category (<0.6 g/kg/day). Lower protein intake was also more common among women with a high school education or less, women with household incomes below the poverty line, unemployed women, and those on government insurance.

Protein intake of 0.8 to less than 1.1 g/kg/day is associated with lower SGA odds

The findings showed that women in the third quartile of protein intake (0.8 to less than 1.1 g/kg/day) had 22% lower adjusted odds of SGA below the tenth percentile than those consuming 1.1 g/kg/day or more (the fourth and highest quartile).

However, there was no consistent dose response across quartiles. Women in the two lowest intake categories also did not have significantly different SGA odds from the highest-intake group, and the spline analysis found no evidence of a nonlinear association between protein intake and SGA.

The observed SGA proportions were also similar before adjustment, at 10.5% in the third quartile and 10.8% in the highest quartile. In addition, the association was not reproduced in a sensitivity analysis that used absolute daily protein intake and excluded BMI from the model. These findings, therefore, do not identify an optimal intake range or establish that changing protein intake would prevent SGA.

The association with SGA is consistent with earlier research. Earlier animal studies have suggested nonlinear and sex-specific effects of maternal protein intake on offspring health, but human studies are essential for understanding maternal and fetal outcomes.

The authors proposed several possible mechanisms, including improved placental blood flow through nitric oxide production from L-arginine, placental angiogenesis, enhanced maternal-fetal nutrient exchange, and mechanistic target of rapamycin (mTOR) signaling, but these remain speculative.

Some research suggests that dietary patterns high in red and processed meat are associated with increased SGA risk, whereas diets rich in fish, eggs, and plant-based foods may be protective. However, evidence on protein sources remains limited and heterogeneous. Future studies should examine the roles of different protein sources.

No association with LGA, HDP, or GDM

Protein intake was not associated with the risk of gestational diabetes mellitus, hypertensive disorders of pregnancy, or large-for-gestational-age birth in the adjusted quartile analyses. The authors suggested that the relatively low incidence of GDM may have limited statistical power and that analyses by protein source might have produced different findings.

Although spline analyses suggested nonlinear associations with HDP and GDM, these associations were no longer statistically significant in the sensitivity analysis using absolute protein intake without BMI adjustment, suggesting that BMI may have influenced the findings.

Limitations

The sociodemographic composition of the study sample limits its generalizability. The relatively low incidence of GDM may have underpowered the study to detect associations between GDM and protein intake.

The researchers did not assess changes in diet and nutrition over the course of pregnancy, protein sources (including animal versus plant-based), dietary patterns, the frequency and quantity of specific protein foods, the use of supplements, or alternative nutrition sources.

Recall error is possible due to reliance on a food frequency questionnaire. BMI calculations were also not age-specific for adolescent participants, which may have led to underestimation or overestimation.

Expressing protein intake relative to body weight may have artificially altered the results. Adjusting for BMI may also have attenuated, erased, or reversed associations. The null sensitivity analysis using absolute protein intake without BMI adjustment reinforces this concern.

The analyses did not include a correction for multiple comparisons, so the isolated significant result should be interpreted cautiously. Residual confounding cannot be excluded either. These should therefore be considered exploratory findings that require validation in future research.

Conclusion

The researchers concluded that maternal protein intake varied widely across demographic and socioeconomic groups. However, only intake of 0.8 to less than 1.1 g/kg/day was associated with lower adjusted odds of SGA than intake of at least 1.1 g/kg/day.

The authors emphasized that the finding was exploratory because there was no dose-response pattern, and supporting analyses were inconsistent. As an observational secondary analysis, the study cannot establish that modifying protein intake would reduce SGA risk, and the association should be confirmed in future studies.