The association between first trimester omega-3 fatty acid supplementation and fetal growth trajectories

Abstract

Background:

Prenatal omega-3 fatty acid supplementation, particularly docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA), has been associated with greater birthweight in clinical trials; however, its impact on fetal growth throughout gestation is unknown.

Objective:

To examine the association between first trimester DHA/EPA supplementation and growth trajectories of estimated fetal weight and specific fetal biometrics measured longitudinally from second trimester to delivery.

Study Design:

In a multisite, prospective cohort of racially diverse, low-risk pregnant women, we used secondary data analysis to examine fetal growth trajectories in relation to self-reported (yes/no) first trimester DHA/EPA supplementation. Fetal ultrasonographic measurements including, abdominal circumference (AC), biparietal diameter (BPD), femur length (FL), head circumference (HC), and humerus length (HL), were measured at enrollment (8–13 weeks) and at up to five follow-up visits. Estimated fetal weight (EFW) and HC:AC ratio (a measure of growth symmetry) were calculated. Fetal growth trajectories were modeled for each measure using a linear mixed model with cubic splines. If significant differences in fetal growth trajectories between groups were observed (global p<0.05), we performed weekly comparisons to determine when in gestation these differences emerged. Analyses were adjusted for maternal sociodemographics, parity, infant sex, total energy consumption, and diet quality score. We repeated all analyses using estimated dietary DHA/EPA intake, dichotomized at the recommended cutoff for pregnant and lactating women (≥ 0.25 g/day vs. <0.25 g/day), among women who did not report supplement intake in the 1^st trimester.

Results:

Among 1535 women, 143 (9%) reported DHA/EPA supplementation in the first trimester. Overall, first trimester DHA/EPA supplementation was associated with statistically significant differences in fetal growth trajectories during pregnancy. Specifically, EFW was larger among women with DHA/EPA supplementation compared to those without supplementation (global p=0.028) with significant weekly differences in median EFW most apparent between 38 to 41 weeks of gestation (median EFW difference at 40 weeks = 114 grams). Differences in fetal growth trajectories for AC (p=0.003), HC (p=0.003), and HC:AC ratio (p=0.0004) were also identified by supplementation status. In weekly comparisons, DHA/EPA supplement use was associated with larger median AC (changed from 2 to 9 mm) in mid-pregnancy onward (19 to 41 weeks), larger median HC between 30 to 33 weeks, and smaller median HC:AC ratio in the 2^nd and 3^rd trimesters of pregnancy. There were no specific weekly differences in fetal FL or HL by DHA/EPA supplementation. First trimester dietary sources of DHA/EPA among women with no first trimester DHA/EPA supplementation (n=1392) were associated with differences in fetal BPD (p=0.043), but not other metrics of fetal growth. At the recommended dietary DHA/EPA levels compared to below recommended levels, BPD was larger between 38 to 41 weeks of gestation.

Conclusion:

In this racially diverse pregnancy cohort, first trimester DHA/EPA supplementation was associated with significant increases in fetal growth, specifically greater estimated fetal abdominal circumference in the 2^nd and 3^rd trimesters of pregnancy.

Introduction

The major dietary source of omega-3 polyunsaturated fatty acids (PUFAs), particularly docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA) is seafood. Yet, many pregnant women in the U.S. do not meet the recommended intake of 8–12 ounces per week, ^1,2 due partly to limited fish intake given the concerns of environmental toxins including mercury. ^3–5 As an alternative, some pregnant women take DHA/EPA supplements, which is reported to be 7% according to the National Health and Nutrition Examination Survey 2003–2014 (NHANES). ^1,6

There is growing evidence from epidemiologic studies that prenatal DHA/EPA intake is associated with longer gestation and higher birth weight, although its impact on size-for-gestational-age and neonatal anthropometry is mixed. ^7–9 In a 2018 Cochrane review of 70 randomized control trials (RCTs), maternal consumption of DHA/EPA (from supplements or dietary sources) during pregnancy compared with placebo or no intake was associated with longer mean gestation, and decreased risk of low birth weight, but no difference in risk of small- or large-for-gestational age (SGA, LGA), and inconsistent differences in neonatal length and head circumference.⁷

There is a limited number of studies examining the association between DHA/EPA consumption and fetal growth as measured during pregnancy with conflicting results and studies have been in Northern European populations with higher fish consumption than in the U.S. ^8,10,11 Moreover, while dietary fish consumption is the largest source of EPA/DHA, data gaps remain regarding supplementation in relation to fetal growth. Therefore, the primary aim of this study was to investigate associations between maternal first trimester DHA/EPA intake and longitudinal trajectories of fetal biometric measures and weight in a diverse pregnancy cohort in the US and to determine when during pregnancy fetal growth may start to deviate.

Material and Methods

Study population

The Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Fetal Growth Studies – Singletons (2009–2013) was a prospective cohort of 2802 racially diverse, pregnant women from 12 clinical sites across the United States with the objective of establishing standards for fetal growth among 4 self-identified racial/ethnic groups (non-Hispanic White, non-Hispanic Black, Hispanic, Asian/Pacific Islanders).¹² Complete description of the study design has been published.¹³ Briefly, women were enrolled between 8w0d to 13w6d of gestation and randomized to one of four follow-up visit schedules for a total of five follow-up visits. At each visit women were interviewed by trained research personnel on their demographic, health, and lifestyle characteristics. Initially the study collected dietary information only from women randomized to one of the four follow-up groups, but after procurement of additional funding, dietary data collection was expanded to all subsequently enrolled participants. ¹⁴ Hence, only a portion of participating women (n=1980) were asked to report their dietary intake in the past three months. In a secondary data analysis using the sample of women with complete dietary intake data (n=1699), we excluded women who exited from the study (n = 23), were missing DHA/EPA supplementation data (n=61), ultrasound data (n=6), covariates (n=1), or had implausible dietary data¹⁵ (<600 or >6000 kcal/d) (n=73). Thus, the analytic sample was comprised of 1535 women. All women provided informed consent prior to participation and human subject approval was obtained at all participating sites, data coordinating center, and NICHD.

Measurements

Outcome

The primary outcome was the longitudinal trajectory of fetal growth. Fetal growth was measured at each research visit at 16–22, 24–29, 30–33, 34–37, and 38–41 weeks of gestation. Ultrasounds were conducted by trained sonography experts using standard operating procedures and identical equipment¹⁶ to measure fetal head circumference (HC), biparietal diameter (BPD), abdominal circumference (AC), femur length (FL), and humerus length (HL). Quality control was performed on a 10% random sample of ultrasound images. Blank images of fetal biometrics (BPD, HC, AC, and FL) were remeasured by an expert sonographer. The correlation between the expert reviewer and site sonographers was 0.91–0.99 for all measures across visits, with the majority having a correlation ≥ 0.95 indicating excellent reliability. There were low coefficients of variation (≤3%) between the sonography experts and primary site sonographers for each fetal biometric overall and by study visit indicating low variability. Further details regarding the quality control protocol along with the mean differences detected between site sonographers and expert sonographer have been published elsewhere.¹⁶ Estimated fetal weight (EFW) was calculated, ¹⁷ and the ratio of HC:AC was computed as a measure of fetal growth symmetry. ^18–20

Secondary outcomes were extracted from maternal and neonatal medical records at delivery and included birthweight and date of delivery. Gestational age was calculated as the difference between the date of delivery and date of last menstrual period, which was confirmed by ultrasound at enrollment. Newborns were classified as preterm (< 37 weeks gestation), SGA (defined as below the 10^th percentile) and LGA (defined as above the 90^th percentile). ²¹

Exposure

At enrollment, women provided a list of supplements they used since becoming pregnant. The Slone Drug Dictionary was used to code and classify the free text supplement data based on their ingredients.²² All reported brand names and generic forms of fish oil and other products containing DHA, EPA, DHA/EPA combined, DHA + Vitamin D, and multivitamin + DHA were categorized into the DHA/EPA supplementation group as a binary variable (yes/no). DHA/EPA supplementation information only from the first trimester was used to conduct prospective analysis.

The mean dietary DHA/EPA consumption (grams/day) in the periconception and first trimester was computed using information from women’s self-administered food frequency questionnaires (FFQ).¹⁴ The FFQ was a modified version of the Diet History Questionnaire-II ²³ and collected information on how often women consumed each food and their portion size in the past three months. The analysis of the FFQ data was completed using the Diet*Calc software developed by the National Cancer Institute (NCI, Bethesda, MD). Women who reported no first trimester supplementation were categorized based on whether they met the recommended intake of ≥ 0.25 g/d of dietary DHA/EPA, ^24,25 since women who took supplements should have met the daily requirements. This categorization was created to identify whether dietary sources, apart from supplementation are sufficient to impact fetal growth.

Covariates

Information on women’s race/ethnicity (non-Hispanic White (ref), non-Hispanic Black, Hispanic, Asian/Pacific Islander), age, parity (nulliparous (ref), multiparous), education (< high school, high school diploma/GED, some college, Bachelor’s degree, Master’s degree or higher (ref)), pre-pregnancy weight and height, and infant sex (male (ref), female) were self-reported or abstracted from medical records. Pre-pregnancy body mass index (BMI) was calculated from self-reported weight and height. Total energy intake and women’s diet quality based on the modified version of the Healthy Eating Index 2010 (HEI-2010) excluding PUFAs, were estimated from the FFQ.²⁶

Analysis

Baseline maternal characteristics were compared using chi-square or Fisher exact tests for categorical and student t-tests for continuous variables. To make the distribution conform to the assumption of normality, biometric measures and EFW were log transformed. For each biometric measurement, the combined effect of time (gestational age in weeks) and exposure (first trimester DHA/EPA supplementation) on fetal growth trajectories in the second trimester onward were estimated by fitting a linear mixed model. Time was modeled with cubic splines for the fixed, and a cubic polynomial for the random effects in gestational age with three knots chosen at 25^th, 50^th, and 75^th percentiles of gestational ages that evenly split the distribution. The total association of exposure was evaluated as the sum of main exposure terms and interaction between exposure and gestational age. For each biometric measure, global differences in the longitudinal trajectory were determined between the two exposure groups. Where the global difference was significant, weekly pairwise comparisons were performed. In a sensitivity analysis, we applied the Benjamini-Hochberg (B-H) procedure to account for multiple comparisons.²⁷ To examine the association between dietary DHA/EPA and fetal growth, analyses were repeated restricting the sample to women who had information on dietary DHA/EPA intake and did not report first trimester DHA/EPA supplementation (n= 1392). Linear regression and logistic regression models were also fit to examine the associations between first trimester supplementation and birthweight, gestational age at delivery, preterm delivery, SGA and LGA in order to put our findings into context with the literature. All analyses were adjusted for covariates.

We repeated the analysis among term pregnancies only (n=1445) to ensure that the potential differences in growth between preterm and term infants did not affect the association. Furthermore, since dietary information was available on a subset of women, we performed a sensitivity analysis to ensure the robustness of our findings in the full cohort of women (n=2606) with supplementation information without considering dietary intake. All analyses were conducted in SAS version 9.4 (version 9.4, SAS Institute, Inc., Cary, North Carolina) and R (version 3.6.1, http://www.R-project.org).

Results

Among 1535 participants in the study, 9% (n=143) reported taking supplements containing DHA/EPA in the first trimester. Participants who took DHA/EPA supplements were more likely to be older, non-Hispanic White, more educated, nulliparous, and have better diet quality (higher HEI score) and a lower total energy consumption than participants who did not take DHA/EPA supplements. The median dietary DHA/EPA intake among study participants was 0.17 g/d (interquartile range= 0.29 g/d) and 3% of women (n=44) did not consume any dietary DHA/EPA. (Table 1)

Table 1.

Women’s characteristics at enrollment

Characteristics	Overall N=1535	First Trimester DHA/EPA Supplementation
		Yes n=143 (9%)	No n=1392 (91%)	P-valuea
Age, years - mean (SD)	28 (6)	31 (4)	28 (6)	<0.0001
Race/ethnicity –n (%)				<0.0001
White, non-Hispanic	320 (21)	79 (55)	241 (17)
Black, non-Hispanic	457 (30)	15 (10)	442 (32)
Hispanic	465 (30)	21 (15)	444 (32)
Asian/Pacific Islanders	293 (19)	28 (20)	265 (19)
Pre-pregnancy obesity status – n (%)				0.168
Non-obese	1279 (83)	125 (87)	1154 (83)
Obese	256 (17)	18 (13)	238 (17)
Education – n (%)				<0.0001
< High School	176 (11)	4 (3)	172 (12)
High school diploma/GED	303 (20)	4 (3)	299 (21)
Some college	474 (31)	28 (20)	446 (32)
Bachelor’s degree	335 (22)	49 (34)	286 (21)
Masters or advanced degree	247 (16)	58 (41)	189 (14)
Parity – n (%)				<0.0001
Nulliparous	701 (46)	90 (63)	611 (44)
Multiparous	834 (54)	53 (37)	781 (56)
Dietary DHA/EPA, g/d – median	0.17	0.17	0.17	0.74
Dietary DHA/EPA, g/d – n (%)				0.73
< 0.25 g/d	978 (64)	93 (65)	885 (64)
≥ 0.25 g/d	557 (36)	50 (35)	507 (36)
HEI 2010 Total Scoreb - mean (SD)	55 (10)	60 (8)	54 (10)	<0.0001
Total energy consumption, kcal/d - median	1897	1821	1914	<0.0001

Abbreviations. FA: fatty acid; GED: general education development; g: grams; g/d: grams per day; HEI: Healthy Eating Index; kcal: kilo calories; SD: standard deviation.

^bTotal scores excluded the PUFAs

^ap-values were obtained using x² tests of independent or Fisher exact test for categorical variables and t-tests or one-way ANOVA for continuous variables.

The overall changes in the fetal growth biometrics with respect to first trimester DHA/EPA supplementation are summarized in Table 2. Trajectories of growth are also presented in Figure 1 for EFW and Supplementary Figure 1 for AC, BPD, FL, HC, HL, and HC:AC ratio. Significant adjusted global differences in fetal growth trajectories between women who did and did not take DHA/EPA supplements were detected for EFW (P = 0.028), HC (P = 0.003), AC (P = 0.003), and HC:AC ratio (P = 0.0004). The median EFW was consistently higher in women reporting DHA/EPA supplementation compared to women without, starting at week 19, with the differences becoming statistically significant toward the end of pregnancy when the EFW was 83 to 134 grams higher in women with DHA/EPA supplementation between 38 to 41 weeks of gestation (Figure 1 and Table 3). Similarly, week-by-week analyses of biometric measures which were observed to have significant global differences in trajectories suggested specific time points where differences were most apparent; that is, a 1.64 to 1.83 mm significantly larger fetal HC between weeks 30–33 of gestation, 1.65 to 9.47 mm significantly greater fetal AC between weeks 19–41 of gestation, and a smaller HC:AC ratio in mid (17–24 weeks) and late (38–41 weeks) pregnancy in women with first trimester DHA/EPA supplementation compared to women without supplementation after adjusting for covariates (Supplementary Table 1). After the B-H adjustment, the statistically significant weekly differences observed in EFW and HC by DHA/EPA supplementation were attenuated and only the weekly median AC and HC:AC ratio remained significantly different between the two supplementation groups. We repeated the global and weekly comparisons tests for the differences in fetal growth outcomes by first trimester DHA/EPA supplementation excluding preterm births and found overall similar results (Supplementary Table 2).

Table 2.

Summary of changes in fetal growth trajectories and weekly differences in response to DHA/EPA supplementation in the first trimester (n=1535)

Outcome	Trajectory (DHA/EPA supplementation vs. not)	Gestational age for significant weekly pairwise differencesb	Direction of associationc
	Adjusted P-valuea
Abdominal circumference	0.003	19–41	Higher
Biparietal diameter	0.076	29–36	Higher
Estimated fetal weight	0.028	38–41	Higher
Femur length	0.268	-d	-
Head circumference	0.003	30–33	Higher
HC:AC ratio	0.0004	17–24, 38–41	Lower
Humerus length	0.152	38–41	Higher

Abbreviations: AC, abdominal circumference; HC, head circumference

^aValues differed significantly from the no DHA/EPA supplementation group and p-values were obtained by Wald test

^bLinear mixed models were adjusted for maternal age, race/ethnicity (non-Hispanic White (ref), non-Hispanic Black, Hispanic, Asian/Pacific Islanders), pre-pregnancy obesity status (non-obese (ref), obese), education (< high school, high school diploma/GED, some college, Bachelor’s degree, Master’s degree or higher (ref), parity (nulliparous (ref), multiparous), infant sex (male (ref), female), Healthy Eating Index (HEI)-2010 score, and total energy consumption.

^cThe direction of change in the outcome indicates increase or decrease in DHA/EPA supplementation compared to no supplementation.

^dThere were no gestational age for which weekly pairwise comparisons were significantly different

Figure 1.

Median estimated fetal weight by first trimester DHA/EPA supplementation and gestation

Table 3.

Median estimated fetal weight (grams) by first trimester supplementation status and gestation (n=1535)

Gestational age, week	Median (95% CI) estimated fetal weighta, grams
	DHA/EPA supplements	No DHA/EPA supplements
15	111 (108, 113)	111 (110, 112)
16	140 (137, 143)	140 (139, 142)
17	176 (173, 180)	176 (175, 178)
18	220 (215, 224)	219 (217, 221)
19	271 (266, 277)	270 (268, 272)
20	331 (324, 337)	329 (326, 332)
21	399 (391, 407)	396 (392, 399)
22	477 (467, 486)	472 (468, 476)
23	564 (553, 575)	559 (554, 563)
24	662 (650, 675)	655 (650, 661)
25	772 (757, 787)	763 (757, 770)
26	894 (876, 911)	883 (875, 890)
27	1028 (1008, 1049)	1015 (1006, 1024)
28	1177 (1154, 1201)	1162 (1152, 1172)
29	1342 (1316, 1369)	1324 (1312, 1335)
30	1523 (1493, 1554)	1501 (1488, 1514)
31	1719 (1684, 1754)	1692 (1678, 1707)
32	1927 (1887, 1968)	1896 (1879, 1913)
33	2145 (2101, 2191)	2109 (2090, 2128)
34	2369 (2319, 2421)	2326 (2305, 2348)
35	2594 (2538, 2652)	2543 (2520, 2567)
36	2818 (2755, 2882)	2758 (2732, 2784)
37	3039 (2969, 3112)	2969 (2940, 2998)
38	3260 (3180, 3342)b, c	3177 (3146, 3209)
39	3480 (3388, 3575)b, c	3383 (3347, 3420)
40	3703 (3590, 3820)b, c	3589 (3537, 3641)
41	3931 (3782, 4087)b, c	3797 (3885, 3931)

Abbreviation: CI, confidence interval

^aComputed using Hadlock ¹⁷ formula from head circumference, abdominal circumference, and femur length

^bValues differed significantly from the no DHA/EPA supplementation group (weekly pairwise comparisons p-values <0.05)

^cAfter adjustment for multiple comparison the differences became non-significant p-value > 0.1

P-valued were obtained by Wald test adjusted for maternal age, race/ethnicity (non-Hispanic White (ref), non-Hispanic Black, Hispanic, Asian/Pacific Islanders), pre-pregnancy obesity status (non-obese (ref), obese), education (< high school, high school diploma/GED, some college, Bachelor’s degree, Master’s degree or higher (ref), parity (nulliparous (ref), multiparous), infant sex (male (ref), female), Healthy Eating Index (HEI)-2010 score, and total energy consumption.

We also examined differences in fetal growth trajectories by dietary DHA/EPA using the recommended 0.25 g/d cutoff in women who reported no DHA/EPA supplementation (n = 1392) (Table 4). Significant adjusted global differences were only detected in fetal BPD (global P-value= 0.043) in which women consuming 0.25 g/d or higher of dietary DHA/EPA in the first trimester had a greater fetal BPD at the end of pregnancy (38–41 weeks of gestation) compared to women reporting intakes less than 0.25 g/d.

Table 4.

Summary of changes in fetal growth trajectories and weekly differences in response to dietary DHA/EPA consumption in the first trimester (n=1392)

Outcome	Trajectory Dietary DHA/EPA < 0.25 g/d vs. Dietary DHA/EPA ≥0.25 g/d)	Gestational age for significant weekly pairwise differencesb	Direction of associationc
	Adjusted P-valuea
Abdominal circumference	0.234	29–34	Lower
Biparietal diameter	0.043	38–41	Higher
Estimated fetal weight	0.333	-d	-
Femur length	0.196	-	-
Head circumference	0.414	-	-
HC:AC ratio	0.053	25–38	Higher
Humerus length	0.480	-	-

Abbreviations: AC, abdominal circumference; HC, head circumference

^aValues differed significantly from the no DHA/EPA supplementation group and p-values were obtained by Wald test

^bLinear mixed models were adjusted for maternal age, race/ethnicity (non-Hispanic White (ref), non-Hispanic Black, Hispanic, Asian/Pacific Islanders), pre-pregnancy obesity status (non-obese (ref), obese), education (< high school, high school diploma/GED, some college, Bachelor’s degree, Master’s degree or higher (ref), parity (nulliparous (ref), multiparous), infant sex (male (ref), female), Healthy Eating Index (HEI)-2010 score, and total energy consumption.

^cThe direction of change in the outcome indicates increase or decrease in DHA/EPA supplementation compared to no supplementation.

^dThere were no gestational age for which weekly pairwise comparisons were significantly different

Birthweight was on average 97 g higher (P = 0.04) in women with DHA/EPA supplementation than in those without supplementation in the first trimester (Supplementary Table 3), but associations were not statistically significant after adjustment for covariates, 95% CI: −60.6, 132.3 (Table 5). Average gestational age at delivery also did not differ between the groups (39.4 weeks vs. 39.1 weeks gestation) by supplementation status (P> 0.05). Prevalence of preterm delivery, SGA, and LGA in the sample were 6%, 9%, and 8%, respectively (Supplementary Table 3), which also did not differ significantly by supplementation status (Table 5). These findings remained similar when we examined the association in the larger cohort of women (n=2606) who reported supplement use in the first trimester without adjustment for dietary quality and total energy intake (data not shown). Similarly, there were no significant associations between dietary DHA/EPA consumption and length of gestation, preterm birth, and LGA. However, consuming 0.25 mg/d or more of dietary DHA/EPA was associated with decreased odds of being SGA (aOR = 0.62, 95% CI = 0.52, 0.74).

Table 5.

Adjusted associations between first trimester DHA/EPA supplementation and dietary DHA/EPA and birth characteristics

	Birthweight		GA at delivery		Preterm birth		SGA		LGA
DHA/EPA supplement (n=1535)	Beta	95% CI	Beta	95% CI	OR	95% CI	OR	95% CI	OR	95% CI
Yes	35.87	(−60.59, 132.33)	0.18	(−0.18, 0.53)	0.50	(0.19, 1.30)	1.27	(0.63, 2.54)	0.79	(0.42, 1.50)
No	Ref.		Ref.		Ref.		Ref.		Ref.
Dietary DHA/EPAa (n=1392)
≥ 0.25 g/d	4.23	(−19.94, 28.40)	0.01	(−0.07, 0.09)	0.81	(0.66, 1.01)	0.62	(0.52, 0.74)	0.85	(0.71, 1.01)
< 0.25 g/d	Ref.		Ref.		Ref.		Ref.		Ref.

Abbreviations: CI: confidence interval; LGA, large for gestational age defined as greater than the 90^th percentile; OR: odds ratio; Ref: reference; SGA, small for gestational age defined as less than the 10^th percentile.

^aWomen reporting intake of DHA/EPA supplements were excluded from Dietary DHA/EPA sample.

Preterm birth was defined as gestational age less than 37 weeks at delivery

Models were adjusted for maternal age, race/ethnicity (non-Hispanic White (ref), non-Hispanic Black, Hispanic, Asian/Pacific Islanders), pre-pregnancy obesity status (non-obese (ref), obese), education (< high school, high school diploma/GED, some college, Bachelor’s degree, Master’s degree or higher (ref), parity (nulliparous (ref), multiparous), infant sex (male (ref), female), Healthy Eating Index (HEI)-2010 score, and total energy consumption.

Comment

Principle findings

In this prospective cohort of racially diverse pregnant women, 9% reported taking DHA/EPA supplements in the first trimester, which was consistent with estimates among U.S. pregnant women of 7%. ¹ First trimester DHA/EPA supplementation was associated with increased fetal AC growth starting at 19 weeks and HC growth at 30 weeks of gestation, with a corresponding EFW that was higher across pregnancy . First trimester DHA/EPA supplementation was not associated with differences in the length of gestation, the odds of preterm birth, SGA or LGA.

Consumption of 0.25 g/d or higher of dietary DHA/EPA was associated with significant differences in fetal growth trajectories only for fetal BPD. Nevertheless, consumption of 0.25+ g/d of dietary DHA/EPA compared to less than 0.25 g/d was associated with a significantly lower risk for SGA.

Results in the context of what is known

Our findings that first trimester DHA/EPA supplementation was associated with increased EFW growth, manifested mainly by HC and AC, are generally consistent with prior literature reporting increased birthweight and reduced odds of low birth weight. ^7–9,28 Some studies reporting birthweight differences by omega-3 supplementation during pregnancy attributed these differences to longer gestation.⁹ For instance, in a randomized trial, Carlsen et al. (2013) concluded that the benefits of fish oil may lie with lengthening gestation rather than fetal growth.¹⁰ That study, however, did not have information on fetal biometry past 24 weeks’ gestation. In our study, we found no difference in the length of gestation by DHA/EPA supplementation status while the estimated fetal weight was 97 grams higher in women with DHA/EPA supplementation at 39 weeks of gestation. Our null findings for length of gestation, preterm birth, SGA, and LGA are in line with null findings from several randomized trials of omega-3 supplementation and other observational studies.^9,11,29–32 Taken together, these findings suggest that the increased birthweight associated with DHA/EPA supplementation observed in prior studies is more likely to be due to enhanced biometric growth trajectories rather than solely a factor of longer gestation. After adjustment for multiple comparisons, the results for AC and HC:AC remained significant after adjustment; however, there were no significant differences in EFW in the weekly comparisons of the two groups. The non-significant results for the EFW after multiple comparison correction does reconcile with the results for birthweight in other studies. Given the unadjusted differences identified in this exploratory analysis and small but systematic departure in growth over consecutive weeks, further investigation is warranted.

The evidence for individual fetal biometrics is not consistent. In our study, we found a significant increase in fetal AC in the second and third trimesters and a non-significant increasing trend in fetal HC in the third trimester associated with first trimester DHA/EPA supplementation. In contrast to our findings, in the Generation R cohort of 3380 pregnant women, there was no association between maternal self-reported total fish consumption (grams/week) in the first trimester and fetal head circumference by ultrasound in the second and third trimesters after adjusting for maternal characteristics. ¹¹ Some studies have found increased HC during pregnancy and at birth in response to DHA/EPA consumptions, ^8,33 while others reported no difference. ^10,34–38 The potential reasons for these differences are unknown but may be due to population differences.

Furthermore, our finding of a decreasing trend for fetal HC:AC ratio during pregnancy with a lower ratio observed in the second trimester (17–24 weeks) and at the end of pregnancy (38–41 weeks) in women with DHA/EPA supplementation indicates that while both the HC and AC were each positively associated with supplementation, they did not increase uniformly across pregnancy. The increase in HC fell behind that of the increase in AC at certain points in pregnancy primarily due to the more persistent association with increased AC from supplementation. Since the HC:AC ratio is a measure of asymmetric intrauterine growth, future studies are warranted to further explore the implications of this association.¹⁸

Research implications

The present study is one of the few studies that examined prenatal DHA/EPA consumption and longitudinal fetal growth. Most studies previously examined birthweight as proxy of fetal growth,²⁸ making it difficult to compare the findings. Overall, findings indicate a positive non-significant association between first trimester maternal DHA/EPA supplementation and estimated fetal weight, with increased measures of AC being most consistently apparent to the end of pregnancy. Although these are small differences of uncertain clinical significance, the increasing trends in fetal growth, especially the increase in EFW and AC in response to DHA/EPA supplementation, generate interesting hypotheses. Future studies with long-term follow ups are needed to examine the reproducibility of our results and to further explore the association between prenatal DHA/EPA supplementation with fetal and child growth. These studies will need to focus on the timing, dose, and duration of supplementation in order to better understand the short- and long-term clinical implications of DHA/EPA supplementation.

Clinical implications

Previous studies have found that low fish consumption in the first and second trimester increased chance of preterm birth and a shorter mean gestation.³⁹ Other studies have also shown that DHA supplementation during pregnancy was associated with an increase in birthweight among low-income women living in urban areas.⁴⁰ These findings, together with our finding of increase in trajectories of estimated fetal weight, head circumference and abdominal circumference in response to DHA/EPA supplementation suggest that prenatal supplementation of DHA/EPA starting in the first trimester may be associated with improved fetal growth and pregnancy outcomes, especially among pregnant women with limited access to diets enriched in DHA/EPA. However, given the observational nature of this study, causality cannot be known, and further study is required.

Strengths and limitations

The prospective design of this study and ultrasound collection by trained sonographers at multiple points during pregnancy allowed for an accurate estimation of fetal biometric measures throughout pregnancy. Also, availability of information on both dietary and supplementation of DHA/EPA enabled us to examine the independent association of each source with fetal growth trajectories separately while accounting for potential confounding of other factors such as diet quality and total energy intake.

Our study was limited by lack of information on dose, frequency and duration of supplement intake prior to or during pregnancy. We lacked power to account for multiple comparisons so our findings should be considered exploratory. Since only 9% of women reported taking DHA/EPA supplements, while consistent with previous US reports,¹ statistical power was limited to detect weaker differences, especially for weekly analyses. Similarly, the low median (0.17 g/d) levels of dietary DHA/EPA consumption in our sample may have limited the ability to detect weak associations, especially in weekly comparisons, potentially explaining inconsistency in findings between DHA/EPA consumption from supplementation and diet. Although we have adjusted for various covariates in the regression models, some of the differences in fetal growth trajectories may be attributed to the overall differences in the lifestyle of women with DHA/EPA supplementation compared to those without. Furthermore, due to the small magnitude of differences detected we are unable to make conclusions about the clinical implications of the findings. However, the positive trends observed in fetal growth can inform future studies.

Conclusion

The findings of the present study indicate that in a racially diverse, low-risk pregnancy cohort, DHA/EPA supplementation in the first trimester was associated with increased fetal growth, specifically higher measures of estimated fetal weight, abdominal circumference, and head circumferences in the second and third trimesters. Further research on the long-term implications of the findings on growth and development is warranted.

AJOG at a glance.

1. Why was the study conducted?

   • Some clinical trials have identified increased birthweight associated with prenatal consumption of omega-3 fatty acids.
   • The timing of alteration in fetal growth associated with omega-3 supplementation is unknown.
2. What are the key findings?

   • First trimester greater docosahexaenoic acid/eicosapentaenoic acid (DHA/EPA) supplementation was associated with increased fetal growth in the second and third trimesters, most specifically larger fetal abdominal circumference and smaller HC:AC ratio.
   • First trimester DHA/EPA supplementation was not associated with length of gestation, preterm birth, small- or large-for gestational age birthweight.
3. What does this study add to what is already known?

   • First trimester DHA/EPA supplementation was associated with increased fetal growth starting as early as 19 weeks of gestation.
   • Increased fetal growth associated with first trimester DHA/EPA supplementation was primarily manifested by larger fetal abdominal circumferences, but not long bones.