High-Dose Docosahexaenoic Acid in Newborns Born at Less Than 29 Weeks’ Gestation and Behavior at Age 5 Years: Follow-Up of a Randomized Clinical Trial

Jacqueline F Gould; Rachel M Roberts; Peter J Anderson; Maria Makrides; Thomas R Sullivan; Robert A Gibson; Andrew J McPhee; Lex W Doyle; Jana M Bednarz; Karen P Best; Gillian Opie; Javeed Travadi; Jeanie L Y Cheong; Peter G Davis; Mary Sharp; Karen Simmer; Kenneth Tan; Scott Morris; Kei Lui; Srinivas Bolisetty; Helen Liley; Jacqueline Stack; Carmel T Collins

doi:10.1001/jamapediatrics.2023.4924

. 2023 Nov 20;178(1):45–54. doi: 10.1001/jamapediatrics.2023.4924

High-Dose Docosahexaenoic Acid in Newborns Born at Less Than 29 Weeks’ Gestation and Behavior at Age 5 Years

Follow-Up of a Randomized Clinical Trial

Jacqueline F Gould ^1,^2,^3,^✉, Rachel M Roberts ³, Peter J Anderson ⁴, Maria Makrides ^1,², Thomas R Sullivan ^1,⁵, Robert A Gibson ^1,⁶, Andrew J McPhee ⁷, Lex W Doyle ⁸, Jana M Bednarz ¹, Karen P Best ^1,², Gillian Opie ⁹, Javeed Travadi ^10,¹¹, Jeanie L Y Cheong ¹², Peter G Davis ¹², Mary Sharp ¹³, Karen Simmer ¹⁴, Kenneth Tan ¹⁵, Scott Morris ¹⁶, Kei Lui ¹⁷, Srinivas Bolisetty ¹⁸, Helen Liley ¹⁹, Jacqueline Stack ²⁰, Carmel T Collins ^1,²

¹SAHMRI Women and Kids, South Australian Health and Medical Research Institute, North Adelaide, South Australia, Australia

²Discipline of Paediatrics, Faculty of Health and Medical Sciences, The University of Adelaide, North Terrace Adelaide, South Australia, Australia

³School of Psychology, Faculty of Health and Medical Sciences, The University of Adelaide, North Terrace Adelaide, Adelaide, South Australia, Australia

⁴Turner Institute for Brain and Mental Health, School of Psychological Sciences, Monash University, Melbourne, Victoria, Australia

⁵School of Public Health, Faculty of Health and Medical Sciences, The University of Adelaide, North Terrace Adelaide, South Australia, Australia

⁶School of Agriculture, Food and Wine, The University of Adelaide, Waite Campus, Glen Osmond, South Australia, Australia

⁷Neonatal Medicine, Women’s and Children’s Hospital, North Adelaide, South Australia, Australia

⁸Department of Obstetrics and Gynaecology, The Royal Women’s Hospital, Parkville, Melbourne, Victoria, Australia

⁹Neonatal Services, Mercy Hospital for Women, Heidelberg, Melbourne, Victoria, Australia

¹⁰Newborn Services, John Hunter Children’s Hospital, New Lambton Heights, New South Wales, Australia

¹¹Neonatal Intensive Care Unit, Department of Paediatrics, Te Whatu Ora Waikato, Waikato Hospital, Hamilton, New Zealand

¹²Neonatal Medicine, The Royal Women’s Hospital, Parkville, Melbourne, Victoria, Australia

¹³King Edward Memorial Hospital, Perth, Western Australia, Australia

¹⁴The University of Western Australia, Perth, Western Australia, Australia

¹⁵Department of Paediatrics, Monash University, Monash Children’s Hospital, Clayton, Victoria, Australia

¹⁶Neonatal-Perinatal Medicine, Flinders Medical Centre, Flinders Drive, Bedford Park, South Australia, Australia

¹⁷School of Clinical Medicine, Discipline of Paediatrics and Child Health, University of New South Wales, Sydney, New South Wales, Australia

¹⁸Royal Hospital for Women, Randwick, New South Wales, Australia

¹⁹Mater Research – The Faculty of Medicine, The University of Queensland, South Brisbane, Queensland, Australia

²⁰Neonatal Intensive Care Unit, Liverpool Hospital, Elizabeth, Liverpool, New South Wales, Australia

Accepted for Publication: September 14, 2023.

Published Online: November 20, 2023. doi:10.1001/jamapediatrics.2023.4924

^✉

Corresponding Author: Jacqueline F. Gould, PhD, SAHMRI Women and Kids, South Australian Health and Medical Research Institute, 72 King William Rd, North Adelaide, SA 5007, Australia (jacqueline.gould@sahmri.com).

Author Contributions: Dr Sullivan and Ms Bednarz had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Gould, Roberts, Anderson, Makrides, Sullivan, Gibson, McPhee, Best, Simmer, Morris, Bolisetty, Collins.

Acquisition, analysis, or interpretation of data: Gould, Roberts, Anderson, Makrides, Sullivan, Gibson, Doyle, Bednarz, Best, Opie, Travadi, Cheong, Davis, Sharp, Tan, Morris, Lui, Bolisetty, Liley, Stack, Collins.

Drafting of the manuscript: Gould, Sullivan, Best, Bolisetty.

Critical review of the manuscript for important intellectual content: Gould, Roberts, Anderson, Makrides, Sullivan, Gibson, McPhee, Doyle, Bednarz, Best, Opie, Travadi, Cheong, Davis, Sharp, Simmer, Tan, Morris, Lui, Liley, Stack, Collins.

Statistical analysis: Sullivan, Bednarz.

Obtained funding: Gould, Makrides, Gibson, McPhee, Simmer, Collins.

Administrative, technical, or material support: Gould, Roberts, Anderson, Makrides, Sullivan, Gibson, McPhee, Doyle, Best, Cheong, Davis, Sharp, Tan, Morris, Bolisetty, Stack.

Supervision: Gould, Roberts, Anderson, Makrides, Sullivan, Gibson, McPhee, Simmer, Collins.

Conflict of Interest Disclosures: Dr Gould reported grants from Women and Children’s Hospital Foundation and the National Health and Medical Research Council Australia as well as nonfinancial support from NuMega Ingredients (donated study product) during the conduct of the study and grants from Fonterra and honoraria to support conference travel from NuMega Ingredients and Nestle Nutrition Institute outside the submitted work. Dr Roberts reported grants from Women’s and Children’s Hospital Foundation during the conduct of the study. Dr Makrides reported grants from the National Health and Medical Research Council (Project, Investigator, and Centre of Research Excellence) during the conduct of the study and nonfinancial support from NuMega Ingredients (donated study product for the original trial) and grants from Fonterra outside the submitted work; in addition, Drs Makrides and Gibson had a patent for methods and compositions for promoting the neurological development for preterm infants (2009201540) licensed to NuMega Ingredients. Dr Sullivan reported grants from the National Health and Medical Research Council and Women’s and Children’s Hospital Foundation during the conduct of the study. Dr Gibson reported grants from the National Health and Medical Research Council (Project and Centre of Research Excellence) as well as nonfinancial support from NuMega Ingredients (donated study product for the original trial) during the conduct of the study and grants from Fonterra outside the submitted work. Dr McPhee reported grants from the National Health and Medical Research Council Australia during the conduct of the study. Dr Doyle reported grants from the National Health and Medical Research Council of Australia Centre of Research Excellence Grant outside the submitted work. Dr Best reported grants from MS McLeod Post Doctoral Fellowship outside the submitted work. Dr Collins reported grants from Women’s and Children’s Hospital Foundation and the National Health and Medical Research Council as well as nonfinancial support from NuMega Ingredients (donation of study product for original trial) during the conduct of the study. No other disclosures were reported.

Funding/Support: Financial support for the submitted work was from a project grant from the Women’s and Children’s Hospital Foundation, as well as from the National Health and Medical Research Council Australia (1022112 - N3RO for the original trial; 1146806 for the 5-year follow-up).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The contents of the published material are solely the responsibility of the authors and do not reflect the views of the National Health and Medical Research Council.

Data Sharing Statement: See Supplement 3.

Additional Contributions: We thank the families who generously participated in the N3RO trial and follow-up and the N3RO steering committee, investigative team, and research staff, particularly study coordinator Jemima Gore, GradDipHumNutr, South Australian Health and Medical Research Institute. Families were reimbursed AUD $40 per child for their contribution to the study. The coordinator, the steering committee members, and the investigative team received a salary but were not compensated specifically for their role in this study.

^✉

Corresponding author.

PMCID: PMC10660239 PMID: 37983037

Key Points

Question

Does supplementation of docosahexaenoic acid (DHA) to newborns less than 29 weeks’ gestation after birth improve their behavioral functioning in childhood?

Findings

In this follow-up of a randomized clinical trial of an enteral DHA emulsion in 958 infants born at less than 29 weeks’ gestation, there was no evidence for differences in parent-rated behavioral functioning at 5 years’ corrected age.

Meaning

Increasing neonatal DHA supplementation in infants born at less than 29 weeks’ gestation did not influence behavior in early childhood in this study.

This randomized clinical trial evaluates behavior at age 5 years among children who received high-dose docosahexaenoic acid as preterm infants.

Abstract

Importance

Children born at less than 29 weeks’ gestation are at risk of behavioral difficulties. This may be due in part to the lack of transplacental supply of docosahexaenoic acid (DHA), a key fatty acid with structural and functional roles in the brain.

Objective

To determine whether meeting the neonatal DHA requirement through supplementation is associated with improved behavioral functioning of children born at less than 29 weeks’ gestation.

Design, Setting and Participants

This was a follow-up of children from 10 Australian participating centers in a multi-center, blinded, parallel group randomized clinical trial of infants born at less than 29 weeks’ gestation conducted from June 2012 and September 2015, excluding those with additional fatty acid supplementation or major congenital or chromosomal abnormalities. Follow-up took place from August 2018 to May 2021. Parents of surviving children who had not withdrawn from the original trial were invited to complete questionnaires when the child turned 5 years’ corrected age.

Interventions

Infants were randomized to receive daily enteral emulsions providing 60 mg/kg/d of DHA or a soy-oil emulsion (with no DHA) from within the first 3 days of enteral feeding until 36 weeks’ postmenstrual age or discharge home, whichever occurred first.

Main Outcomes and Measures

The primary outcome of this follow-up was parent-rated behavior and emotional functioning as indicated by the Total Difficulties score of the Strengths and Difficulties Questionnaire. Parents also completed questionnaires about their child’s behavioral manifestations of executive functioning, as well as a range of health outcomes to assess potential longer-term side effects of DHA intervention.

Results

Primary outcome data were available for 731 children (76% of 958 surviving eligible children; 361 in the intervention group and 370 in the control group). Of these 731, 452 (47%) were female, and the mean (SD) corrected age at follow-up was 5.4 (0.5) years. Following imputation for missing data, the mean Total Difficulties score was the same in both groups (intervention group, n = 465; mean [SD], 11.8 [6.3]; control group, n = 493; mean [SD], 11.8 [6.0]; mean difference adjusted for sex, gestational age stratum, and hospital, 0.01; 95% CI, −0.87 to 0.89; P = .98). There was no evidence for differences between the groups in any secondary outcomes of behavior, executive functioning, or health.

Conclusions and Relevance

In this follow-up of a randomized clinical trial, enteral DHA supplementation at the equivalent of the estimated in utero dose for infants born at less than 29 weeks’ gestation did not improve behavioral functioning at age 5 years. There were no indications of adverse effects with DHA supplementation.

Trial Registration

Australian New Zealand Clinical Trial Registry: ACTRN12612000503820

Introduction

Children born preterm are at greater risk of developing behavioral problems than children born at term,^1,2 and more than half of surviving children born at less than 30 weeks’ gestation will have a neurobehavioral disability.³ Children born very preterm are also more likely to experience difficulties with high-order cognitive skills, such as executive functioning^4,5; ongoing chronic health conditions, such as asthma, visual impairments, hearing difficulties, and cerebral palsy^6,7,8,9; and poorer health-related quality of life compared with their term-born counterparts.¹⁰

Infants born at less than 30 weeks’ gestation miss much of the last trimester of pregnancy, which is the peak period of accretion of the ω-3 (n-3) long-chain polyunsaturated fatty acid docosahexaenoic acid (DHA).¹¹ DHA is a key fatty acid in the brain but until recently, its role in infant development had not been defined. Early preterm birth deprives infants of the placental supply of DHA, such that they have reduced neural tissue DHA concentration.¹² Insufficient DHA has been hypothesized to contribute to the poorer neurobehavioral outcomes observed in preterm children born at less than 30 weeks’ gestation, such as behavioral problems^1,2 and average intelligence quotient (IQ) scores approximately 12 points lower^2,13,14 compared with outcomes of term-born children. This hypothesis is supported by our recent finding that supplying the estimated in-utero DHA dose in infants born at less than 29 weeks’ gestation improved IQ at 5 years’ corrected age.¹⁵ We wanted to test whether providing DHA to infants born at less than 29 weeks’ also improved behavioral functioning at 5 years’ corrected age. We also assessed other outcomes known to be adversely affected by early preterm birth, such as executive functioning and health, especially respiratory health given that the DHA intervention increased the risk of bronchopulmonary dysplasia (BPD),¹⁶ which may further increase the risk of respiratory illness and asthma in childhood.^6,9

Methods

Study Design

This is a 5-year follow-up of a multi-center, double-blind, parallel-placebo randomized clinical trial of DHA supplementation for infants born at less than 29 weeks’ gestation: the Omega-3 (N-3 Fatty) Acids for Improvement in Respiratory Outcomes (N3RO) trial. The trial is published,¹⁶ as are the results of IQ assessment¹⁵ and the methodology of the 5-year follow-up (Supplement 1).¹⁷ Reporting of methods and results follow the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline. The investigator-designed and -led N3RO trial was approved by the Southern Adelaide Clinical Human Research Ethics Committee (35.12, approved on March 12, 2012) and the Women’s and Children’s Health Network Human Research Ethics Committee (REC2434/12/14, approved on December 16, 2011), and the 5-year follow-up was approved by the Women’s and Children’s Health Network Human Research Ethics Committee (approved on March 8, 2017, HREC/16/WCHN/184). The 5-year follow-up involved parental completion of a survey, hosted on Research Electronic Data Capture (REDCap) software platform.¹⁷ Families were reimbursed with an AUD$40 gift card.

Participants

Infants were originally recruited into the N3RO trial from 13 centers in Australia, New Zealand, and Singapore between June 2012 and September 2015. All surviving children from the 10 Australian sites (eAppendix in Supplement 2) were eligible for this 5-year follow-up, provided they had not withdrawn from the N3RO trial.¹⁷ All children were born at less than 29 weeks’ gestation, recruited within 3 days of their first enteral feed, and had a guardian able to provide written informed consent. Exclusion criteria were participation in another fatty acid intervention, major congenital or chromosomal abnormalities, DHA supplementation over 250 mg per day by a breastfeeding mother, or reception of intravenous lipids containing fish oil.¹⁶ The follow-up commenced August 29, 2018, and the last surveys were completed on May 22, 2021.

Randomization and Blinding

Infants were assigned a unique identification number upon enrollment into the N3RO trial.¹⁶ Randomization used permuted blocks with stratification for gestational age at birth (<27 weeks or 27-28 weeks’ gestation), sex, and center. Infants from a multiple birth were randomized individually. The randomization schedule was generated by an independent statistician and implemented using a secure web-based system.

Emulsions were identical in viscosity, color, and packaging.¹⁶ Families, study staff, and investigators were blinded to group allocation throughout the trial but were able to request knowledge of their group allocation from the independent statistician after completion of the N3RO trial primary outcome analyses for BPD.

Intervention

Randomized infants received a study enteral emulsion 3 times daily immediately before an enteral feed, from within 3 days of their first enteral feed until 36 weeks’ postmenstrual age or discharge home, whichever occurred first. The intervention emulsion provided 60 mg of DHA per kg of body weight per day from fractionated tuna oil, and the control emulsion contained soy oil without DHA. All infants received breastmilk or preterm infant formula containing approximately 20 mg/kg/d DHA once full enteral feeds were reached.

Outcomes and Measures

All outcomes were parent-reported through a survey.¹⁷ The primary outcome of this follow-up study was behavioral and emotional functioning as assessed by the Total Difficulties Score of the Strengths and Difficulties Questionnaire (SDQ). The SDQ requires parents to rate their child’s behavioral symptoms compared with other children of the same age.¹⁸

Secondary outcomes included other scales from the SDQ (Emotional Symptoms, Conduct Problems, Hyperactivity/Inattention, Peer Relationship Problems, Prosocial Behavior, and Impact).¹⁸ Higher scores for SDQ scales (excepting Prosocial Behavior) indicate more perceived symptoms. All scales and indices from the Behavior Rating Inventory of Executive Functioning (BRIEF; preschool edition for children younger than 6 years’ corrected age at time of assessment, and the second edition of the BRIEF for children 6 years’ corrected age and older at the time of survey completion) were used to assess behavioral manifestations of executive function.^19,20 Scores were age standardized, with correction for preterm birth,^21,22 to a mean (SD) of 50 (10), where higher scores indicate more symptoms of executive dysfunction. For exploratory analyses, the SDQ scores were categorized as abnormal or indicative of dysfunction according to Australian norms,²³ and BRIEF scores at or above 65 categorized as indicating a possible dysfunction.¹⁹

Parents also reported whether children had been diagnosed with a cognitive, behavioral, or emotional disorder; attention-deficit/hyperactivity disorder; autism spectrum disorder; or blindness, hearing-loss, deafness, physical disability, or other medical conditions by a health professional. In addition, parents were asked to indicate any respiratory-related hospital admissions or surgical procedures the child received since discharge from the newborn hospitalization, whether the child had received any allied health care services, and whether the child had received any special education support at school (if attending school). The International Study of Asthma and Allergies in Childhood²⁴ questionnaire was included to indicate allergic disease, particularly any respiratory symptoms, as the N3RO trial targeted BPD.¹⁶ The Pediatric Quality of Life Inventory (PedsQL) was used to indicate health-related quality of life, with higher scores indicating higher quality.²⁵ Deaths between the completion of the N3RO trial and the 5-year follow-up were considered as serious adverse events and were reviewed by an independent mortality review committee that was unaware of group allocation.

Sample Characteristics

Characteristics were collected during the original N3RO trial, as well as at the 5-year follow-up. Parents were interviewed at enrollment about demographic information and medical records were reviewed to determine baseline pregnancy and infant characteristics. Postrandomization events during hospitalization were collected by study staff and from medical records in the N3RO trial.

At the 5-year follow-up, the survey questioned the child’s family structure, primary language spoken at home, whether the child attends or attended preschool or full-time school, and child dietary consumption of fish meals and DHA-containing supplements. Given the potential influence of the home and family environment on child behavioral and emotional functioning,^26,27,28 we included 3 parent-completed assessments: the 12-item General Family Functioning scale (considered the short Family Assessment Device) to measure problem solving by the family,²⁹ the Parental Involvement in Developmental Advance subscale of the StimQ to assess parental engagement in child learning activities,³⁰ and the Parenting Scale of dysfunctional parenting in disciplinary situations.³¹

Statistical Analysis

To detect a mean difference of 0.25 SDs (approximately 2 points) between groups in the Total Difficulties score of the SDQ with 90% power (2-tailed α .05), 338 children per group were required (676 total).

Analyses were undertaken on an intention-to-treat basis for all surviving children from the Australian centers, according to a prespecified statistical analysis plan (Supplement 1) using Stata version 17 software (StataCorp). Outcomes of intervention and control group children were compared using generalized linear models, with generalized estimating equations to account for clustering due to multiple births within a family. Continuous, binary, and count outcomes were analyzed using linear, log binomial and negative binomial models, respectively, with adjustment for stratification variables sex, center enrolled, and gestational age (<27 weeks or 27-28 weeks). Preplanned subgroup analyses were performed to test for evidence of treatment effect modification by child sex and gestational age (less than 27 weeks or 27-28 weeks) for all outcomes. We conducted exploratory post hoc comparisons of the proportion of children from each group whose SDQ or BRIEF scores could be classified as abnormal or indicative of dysfunction. Exploratory analyses were performed on the available data (complete-case analysis) and were analyzed using log binomial regression models, with treatment effects expressed as relative risks (DHA/control). No adjustment was made for multiple comparisons, as there was a single primary outcome, and all other analyses were secondary. Missing outcome data were addressed using multiple imputation under a missing at random assumption, with imputation performed separately by treatment group using fully conditional specification.^32,33 Auxiliary variables for imputation included baseline and hospitalization characteristics measured in the original N3RO trial. Group differences were taken to be statistically significant if the P value for the 2-sided comparative test was less than .05.

Results

Of the 1028 children enrolled at the 10 centers participating in the follow-up, families of 76 children were not invited as the child had been withdrawn (n = 6) or died (n = 70), leaving 952 eligible (Figure). There were 3 children who were withdrawn during the follow-up period, 21 whose parents declined to complete the survey, and 120 who were unable to be contacted for follow-up. Although families of 808 children intended to complete the survey, 77 surveys were not completed. There were 731 children (76% of surviving children; 361 in the intervention group and 370 in the control group) with a score for the primary outcome. Of these 731, 452 (47%) were female, and the mean (SD) corrected age at follow-up was 5.4 (0.5) years.

Figure. — DHA indicates docosahexaenoic acid; SDQ, Strengths and Difficulties Questionnaire.

^aThose who withdrew prior to 5 years were not invited to participate in follow-up but were included in imputed analyses.

Characteristics of the randomized groups in the subset in the follow-up were comparable at baseline, during hospitalization, and at 5 years (Table 1; eTables 1 and 2 in Supplement 2). The mean (SD) percentage of prescribed doses administered to participants was 90.8% (11.3%) in both groups (91.0% [11.2%] in the DHA group and 90.6% [11.3%] in the control group) (eTable 3 in Supplement 2). The proportion of children who had physiological BPD in each treatment group was similar to that in the overall trial (Table 1). The mean age at the time of survey completion was similar. There were 18 families (8 in the intervention group [1.7%] and 10 in the control group [2.0%]) who requested knowledge of the intervention after completion of the BPD analyses.

Table 1. Baseline and Postrandomization Characteristics of the Children and Their Parents or Caregivers.

Characteristic^a	Participants, No. (%)
Characteristic^a	Intervention group (n = 465)	Control group (n = 493)
Baseline characteristics
Gestational age, mean (SD), wk	26.8 (1.4)	26.9 (1.5)
<27 wk	227 (48.8)	234 (47.5)
Sex
Female	216 (46.5)	236 (47.9)
Male	249 (53.5)	257 (52.1)
Singleton birth	337 (72.5)	352 (71.4)
Cesarean delivery	279 (60.0)	292 (59.2)
Birthweight, mean (SD), g	933 (233)	935 (228)
Age at randomization, median (IQR), d	3 (2-4)	3 (2-4)
Age at first dose of trial emulsion, median (IQR), d^b	4 (3-5)	4 (3-5)
DHA level in whole blood at randomization, mean (SD) % of total fatty acids^c	2.8 (0.8)	2.7 (0.8)
Maternal age, mean (SD), y	30.7 (5.7)	30.1 (5.9)
Mother completed secondary education, No./total No. (%)^d	330/429 (76.9)	350/458 (76.4)
Father/other parent completed secondary education, No./total No. (%)^d	288/413 (69.7)	302/430 (70.2)
Neonatal postrandomization characteristics
BPD (physiological definition), No./total No. (%)^e	215/439 (49.0)	209/474 (44.1)
BPD (clinical definition)	240 (51.6)	239 (48.5)
Length of hospital stay, median (IQR), d	89 (72-107)	87 (71-108)
Child characteristics at 5 y
Corrected age at time of survey completion, mean (SD), y^f	5.4 (0.5)	5.4 (0.4)
Lives with both parents living together, No./total No. (%)^d	290/357 (81.2)	278/362 (76.8)
Commenced full-time schooling, No./total No. (%)^d	207/357 (58.0)	209/362 (57.7)
General Family Functioning score, mean (SD)^g	1.4 (0.4)	1.5 (0.5)
Parental Involvement in Developmental Advance score, mean (SD)^h	12.8 (2.4)	12.8 (2.6)
Parenting style total score (from Parenting Scale), mean (SD)ⁱ	2.9 (0.6)	3.0 (0.5)

Open in a new tab

Abbreviations: BPD, bronchopulmonary dysplasia; DHA, docosahexaenoic acid.

^{^a}

Maternal, paternal, and family characteristics have been summarized at the infant level (ie, parents or families were counted multiple times if they had multiple infants).

^{^b}

Datum was missing for 1 infant in the intervention group.

^{^c}

Data were missing for 4 infants in the intervention group and 4 infants in the control group.

^{^d}

Information not provided by some caregivers in these areas.

^{^e}

Physiological challenge not completed (7 due to medical advice, 3 due to parent refusal, 4 due to equipment not available, 27 due to staffing issues, and 4 incomplete physiological challenge).

^{^f}

Includes 361 children in the intervention group and 370 children in the control group who had primary outcome data available.

^{^g}

General Family Functioning subscale from the Family Assessment Device measuring structural, organizational, and transactional characteristics of families. Applicable to living situations where there are at least 2 adults living together with the child; assessment not completed by single or sole parents. Scores range from 1 to 4, with higher scores indicating more problematic functioning. Includes 290 children in the intervention group and 274 children in the control group.

^{^h}

Parental Involvement in Developmental Advance subscale of the StimQ measuring cognitive stimulation at home. Scores range from 0 to 15, with higher scores indicating higher parental involvement in developmental activities. Data were missing for 114 children in the intervention group and 139 children in the control group.

^ⁱ

Parenting style total score from the Parenting Scale measuring dysfunctional discipline practices in parents. Scores for parenting style total score range from 1 to 7, with higher scores indicating more dysfunctional parenting. Data were missing for 120 children in the intervention group and 146 children in the control group.

Primary and Secondary Outcomes

Total Difficulties score of the SDQ did not differ between the intervention and control groups (intervention group mean [SD], 11.8 [6.3]; control group mean [SD], 11.8 [6.0]; adjusted mean difference, 0.01; 95% CI, −0.87 to 0.89; P = .98) (Table 2). Other scores from the SDQ likewise did not differ between the randomized groups nor was there any group difference in overall behavioral manifestations of executive functioning indicated by the Global Executive Composite of the BRIEF (intervention group mean [SD], 56.9 [16.6]; control group mean [SD], 57.0 [15.9]; adjusted mean difference, −0.2, 95% CI, −2.51 to 2.12) or other scores from the BRIEF (Table 2).

Table 2. Primary Outcome and Secondary Behavioral Outcomes at 5 Years’ Corrected Age.

Outcome	Mean (SD)		Adjusted mean difference (95% CI)^a	P value
Outcome	Intervention group (n = 465)	Control group (n = 493)	Adjusted mean difference (95% CI)^a	P value
SDQ^b
Total Difficulties score (primary outcome)	11.8 (6.3)	11.8 (6.0)	0.01 (−0.87 to 0.89)	.98
Emotional Symptoms score	2.5 (2.2)	2.3 (2.1)	0.16 (−0.16 to 0.48)	.34
Conduct Problems score	2.4 (2.3)	2.7 (2.2)	−0.26 (−0.58 to 0.06)	.11
Hyperactive/Inattention score	5.0 (2.2)	4.9 (2.3)	0.02 (−0.30 to 0.34)	.90
Peer Relationship Problems score	1.9 (2.0)	1.9 (1.9)	0.03 (−0.25 to 0.32)	.81
Prosocial Behavior score	7.9 (2.1)	8.0 (2.0)	−0.01 (−0.32 to 0.29)	.94
Impact score^c	1.2 (2.2)	1.0 (1.9)	0.12 (−0.18 to 0.41)	.43
BRIEF^d
Inhibit scale T score^e	54.4 (14.0)	54.2 (13.0)	0.16 (−1.80 to 2.11)	.87
Emotional Control scale T score^e	55.0 (15.8)	55.1 (15.0)	−0.20 (−2.39 to 1.99)	.86
Shift scale T score^e	53.6 (12.7)	52.5 (12.1)	1.04 (−0.76 to 2.84)	.26
Working Memory scale T score^f	59.0 (17.0)	58.7 (16.5)	0.13 (−2.36 to 2.63)	.92
Plan/Organize scale T score^e	55.6 (15.7)	55.5 (15.2)	−0.06 (−2.28 to 2.16)	.96
Inhibitory Self-Control Index T score^e	55.2 (15.5)	55.0 (14.4)	0.19 (−2.01 to 2.40)	.86
Flexibility Index T score^e	54.6 (14.8)	54.0 (13.9)	0.42 (−1.68 to 2.53)	.69
Emergent Metacognition Index T score^f	58.2 (17.4)	58.0 (16.6)	−0.01 (−2.49 to 2.47)	.99
Global Executive Composite T score^g	56.9 (16.6)	57.0 (15.9)	−0.20 (−2.51 to 2.12)	.87

Open in a new tab

Abbreviations: BRIEF, Behavior Rating Inventory of Executive Function; SDQ, Strengths and Difficulties Questionnaire.

^{^a}

Mean difference from linear regression model using imputed data and adjusted for randomization strata: sex, gestational age (<27 weeks vs 27-29 weeks), and center.

^{^b}

The SDQ comprises 25 items on psychological attributes equally divided across 5 scales measuring emotional symptoms, conduct problems, hyperactivity/inattention, peer problems, and prosocial behavior and an impact supplement and was administered to parents. The Total Difficulties score is the sum of the emotional scale, conduct scale, hyperactivity/inattention scale, and peer problem scale. Total Difficulties scores range from 0 to 40. Scale scores range from 0 to 10. Higher scores indicate higher risk of clinically significant problems. Missing data were multiply imputed for 104 children in the intervention group and 123 children in the control group.

^{^c}

The Impact score is the sum of 5 items describing the impact of the psychological attributes on the child. Scores range from 0 to 10, with higher scores indicating greater overall distress and impairment.

^{^d}

The BRIEF measures behavioral manifestations of executive function in preschool-aged children and comprises 63 items in 5 clinical scales (Inhibit, Shift, Emotional Control, Working Memory, and Plan/Organize). The clinical scales form 3 summary indices: the Inhibitory Self-Control Index comprising the Inhibit and Emotional Control scales, the Flexibility Index comprising the Shift and Emotional Control scales, and the Emergent Metacognition Index comprising the Working Memory and Plan/Organize scales and 1 composite score (Global Executive Composite). Raw scores were transformed to T scores (mean [SD], 50 [10]) describing a child’s score relative to the scores of the children in the standardization sample. Higher T scores indicate higher levels of problems or difficulties.

^{^e}

Missing data were multiply imputed for 115 children in the intervention group and 135 children in the control group.

^{^f}

Missing data were multiply imputed for 115 children in the intervention group and 136 children in the control group.

^{^g}

Missing data were multiply imputed for 108 children in the intervention group and 134 children in the control group. Observed data includes 8 children for whom the BRIEF2 was completed (7 children in intervention group and 1 child in control group) in place of the BRIEF-Preschool.

There were no group differences in the health-related quality of life as assessed with the PedsQL (Table 3). Responses on the International Study of Asthma and Allergies in Childhood indicated that around 30% of children in both groups had asthma, and there were no differences in the proportion of children with wheezing between groups. A substantial proportion of children from both groups had been hospitalized for respiratory-related conditions hospitalized for respiratory-related conditions (215 of 465 [46.3%] in the intervention group and 243 of 493 [49.6%] in the control group; adjusted risk ratio, 0.95; 95% CI, 0.82 to 1.09) or had undergone a surgical procedure since initial discharge (170 of 465 [36.6%] in the intervention group and 196 of 493 [39.8%] in the control group; adjusted risk ratio, 0.91; 95% CI, 0.77 to 1.09) (Table 3). The numbers of children with a diagnosis of autism spectrum disorder, attention-deficit/hyperactivity disorder, or other behavioral or neurological disorders or physical disabilities were small and did not differ between intervention groups. The number of children who had accessed an allied health service or received special support at school was also similar between the groups.

Table 3. Secondary Health Outcomes at 5 Years’ Corrected Age.

Secondary outcome	No./total No. (%)^a		Adjusted effect (95% CI)^b	P value
Secondary outcome	Intervention group (n = 465)	Control group (n = 493)	Adjusted effect (95% CI)^b	P value
Hearing problem requiring a hearing aid or cochlear implant^c	12/357 (3.4)	4/361 (1.1)	NC	.05
Legally blind or requires glasses or contact lenses^d	67 (14.4)	81 (16.4)	0.89 (0.64 to 1.26)	.52
Diagnosed with ADHD or ADD^e	12/356 (3.4)	11/359 (3.1)	1.10 (0.49 to 2.43)	.82
Diagnosed with autism spectrum disorder^e	26/356 (7.3)	17/359 (4.7)	1.54 (0.90 to 2.63)	.12
Diagnosed with other behavioral disorder^c	8/356 (2.2)	8/359 (2.2)	NC	>.99
Diagnosed with seizures and/or epilepsy^c	6/356 (1.7)	9/359 (2.5)	NC	.60
Diagnosed with cerebral palsy^e	19/356 (5.3)	19/359 (5.3)	1.00 (0.54 to 1.85)	>.99
Diagnosed with intellectual disability^e	15/356 (4.2)	12/359 (3.3)	1.23 (0.60 to 2.52)	.57
Diagnosed with physical disability^e	15/356 (4.2)	12/359 (3.3)	1.24 (0.59 to 2.61)	.57
Any respiratory-related hospital admissions since discharge from hospital^f	215 (46.3)	243 (49.3)	0.95 (0.82 to 1.09)	.46
Diagnosed with other medical conditions^g	119 (25.6)	131 (26.5)	0.94 (0.74 to 1.21)	.64
Any surgical procedures since discharge home from hospital^h	170 (36.6)	196 (39.8)	0.91 (0.77 to 1.09)	.31
Number of respiratory-related hospital admissions since discharge home from hospital, mean (SD)^f	2.0 (4.4)	2.3 (5.4)	0.87 (0.64 to 1.19)	.40
Use of allied health services^g	200 (43.0)	205 (41.5)	1.01 (0.86 to 1.19)	.86
Received special education support at school (among those known to have commenced school)	41/207 (19.8)	46/208 (22.1)	0.87 (0.61 to 1.25)	.46
ISAAC
Wheeze or whistling in the chest everⁱ	262 (56.4)	306 (62.1)	0.91 (0.81 to 1.02)	.10
Wheeze or whistling in the chest in the last 12 mo^g	155 (33.3)	161 (32.6)	1.02 (0.83 to 1.25)	.85
Parent-reported asthma diagnosis ever^g	152 (32.6)	156 (31.5)	1.05 (0.85 to 1.29)	.67
PedsQL, mean (SD)^j
Physical Functioning score^k	80.5 (23.3)	79.8 (22.5)	0.90 (−2.49 to 4.29)	.60
Emotional Functioning score^k	70.9 (19.0)	70.0 (19.3)	0.98 (−1.92 to 3.87)	.51
Social Functioning score^l	76.9 (21.6)	76.3 (21.7)	0.89 (−2.28 to 4.07)	.58
School Functioning score^m	71.1 (21.7)	72.4 (20.5)	−1.13 (−4.13 to 1.87)	.46
Psychosocial Functioning score^l	72.9 (16.9)	72.9 (16.7)	0.22 (−2.24 to 2.68)	.86
Total score^k	75.6 (17.0)	75.2 (16.7)	0.59 (−1.92 to 3.09)	.65

Open in a new tab

Abbreviations: ADD, attention-deficit disorder; ADHD, attention-deficit/hyperactivity disorder; ISAAC, International Study of Asthma and Allergies in Childhood; NC; not calculated owing to small numbers; PedsQL, Pediatric Quality of Life Inventory.

^{^a}

Where missing data were multiply imputed, average numerators across the 100 imputed data sets were rounded to the nearest integer value and hence do not correspond exactly to the reported percentages.

^{^b}

Adjusted effect is the risk ratio from log-binomial regression model, mean difference from the linear regression model, or ratio of means from the negative binomial regression model using unimputed data and adjusted for randomization strata: sex, gestational age (<27 weeks vs 27-29 weeks), and center, unless otherwise indicated.

^{^c}

Adjusted analysis not performed owing to small number of events. P value from Fisher exact test.

^{^d}

Missing data were multiply imputed for 108 children in the intervention group and 132 children in the control group.

^{^e}

Center omitted from adjusted analysis due to nonconvergence or perfect prediction. Adjusted for sex and gestational age only.

^{^f}

Missing data were multiply imputed for 109 children in the intervention group and 132 children in the control group.

^{^g}

Missing data were multiply imputed for 109 children in the intervention group and 134 children in the control group.

^{^h}

Missing data were multiply imputed for 109 children in the intervention group and 132 children in the control group. Log Poisson regression model was used due to convergence issues.

^ⁱ

Missing data were multiply imputed for 109 children in the intervention group and 134 children in the control group. Log Poisson regression model was used due to convergence issues.

^{^j}

The PedsQL is a 23-item instrument measuring health-related quality of life in children with 4 subdomains (Physical Functioning, Emotional Functioning, Social Functioning, School Functioning) and summary scores for Psychosocial Functioning and total score. Scores are linearly transformed to a 0 to 100 scale. Higher scores indicate higher quality of life.

^{^k}

Missing data were multiply imputed for 111 children in the intervention group and 135 children in the control group.

^{^l}

Missing data were multiply imputed for 112 children in the intervention group and 135 children in the control group.

^{^m}

Missing data were multiply imputed for 112 children in the intervention group and 136 children in the control group.

There were 3 deaths (2 in the intervention group and 1 in the control group) in the follow-up period. None of these were deemed related to DHA or study participation by the independent mortality review committee.

Subgroup and Exploratory Analyses

There was no evidence for treatment-effect modification by sex for any outcome (eTables 4 and 5 in Supplement 2). There was some evidence for a treatment-by-gestational age at birth interaction for the BRIEF Flexibility Index score (eTable 6 in Supplement 2) and the number of respiratory-related hospital admissions since discharge home from hospital (adjusted ratio of means, 0.64; 95% CI, 0.41 to 1.00 in children born at less than 27 weeks’ gestation vs 1.23; 95% CI, 0.81 to 1.85 for children born at 27-29 weeks’ gestation) (eTable 7 in Supplement 2). Among children born at less than 27 weeks’ gestation, there was no significant effect of the DHA intervention on the Flexibility Index (eTable 6 in Supplement 2), but in those born at 27 to 29 weeks’ gestation, mean flexibility scores were higher (poorer; adjusted mean difference, 2.82; 95% CI, 0.13 to 5.51). There were no differences between treatment groups with respect to the proportion of children with a SDQ or BRIEF score that could be classified as abnormal or indicative of dysfunction (eTable 8 in Supplement 2).

Discussion

The N3RO trial administered a DHA intervention at the estimated in-utero dose to a large sample of infants born at less than 29 weeks’ gestation and provided the opportunity to test the hypothesis that postnatal DHA intake at the in-utero dose influences behavior at early school age. We assessed multiple aspects of behavioral and emotional functioning, as well as behavioral manifestations of executive functioning, and found that the DHA supplementation had no effect. While we previously found that our DHA intervention for these same infants born at less than 29 weeks’ gestation increased the mean IQ by 3.5 points,¹⁵ behavior is a separate functional domain that may not respond to neonatal DHA supplementation. Similarly, executive functions may not be sensitive to DHA intervention and did not differ in a small subsample of the N3RO trial children assessed as toddlers³⁴ or at age 5 years, as indicated by the working memory, fluid reasoning, or processing speed components of the IQ test.¹⁵

Subgroup analyses by sex and by gestational age at birth were largely null, although among children born at 27 to 28 weeks’ gestation, the average Flexibility Index (ability to shift between activities and modulate emotional response) score of the BRIEF was poorer in the intervention group than control group. A comparable trial of high-dose DHA for breastfeeding mothers of infants born at less than 29 weeks’ gestation also found no overall effects on cognition, motor, language, executive functioning, or behavior but found higher language scores among infants born at less than 27 weeks’ whose mothers had received DHA.³⁵ In our own earlier trial of infants born at less than 33 weeks’ gestation there were a few instances of parent-rated behavior and executive functioning that appeared worse in girls who received extra DHA compared with girls who received the standard dose of DHA, although not in the scales comparable to the Flexibility Index.³⁶ In this study, among children born at 27 to 28 weeks’ gestation, there were fewer with visual impairments in the intervention group than in the control group, consistent with an earlier study³⁷ that found improvement in visual functioning in infants born at less than 33 weeks’ gestation after high-dose DHA supplementation. However, there is a lack of consistency between findings of subgroup analyses between studies. In addition, they are secondary outcomes and need to be interpreted with caution. There is also a lack of support from trials of DHA for more mature preterm infants, pregnant women, or term-born infants for improvements in behavior³⁸ or executive functioning.³⁹

A higher proportion of infants in the N3RO trial, and in this follow-up sample, had experienced BPD if they were randomized to the intervention group.¹⁶ However, the increased risk of BPD with high-dose DHA supplementation was not associated with a change in the benefit to IQ at 5 years’ corrected age.⁴⁰ Results of the current study suggest that DHA supplementation had little impact on health or health-related quality of life at 5 years’ corrected age. Furthermore, there were no differences in wheeze and asthma at 5 years’ corrected age between the randomized groups, providing some reassurance that the DHA-associated increase in risk of BPD observed during the neonatal period was not evident in the longer term. As expected, rates of asthma in both groups were higher than that of term-born children, even compared to those with a family history of allergic disease.⁴¹ Assessment of respiratory and cardiopulmonary functioning in later childhood or adolescence will be important to determine any long-term pulmonary outcomes of the DHA-associated increase in BPD.

Strengths and Limitations

Compliance in N3RO was high (more than 90%),¹⁶ and the method of intervention used in the N3RO trial ensured infants received the full prescribed dose throughout the intervention period. Eligibility for the trial was restricted to a population subgroup of preterm infants particularly at risk of being DHA insufficient⁴² and having a neurobehavioral problem⁴³ due to the degree of prematurity. The study had sufficient power to detect a small effect on the SDQ. Attrition since initial discharge from hospital was 23%, and although we found no evidence of a differential effect by original study arm on attrition, the behavior and health of children lost to follow-up may differ from that of those who participated. Outcomes were parent-reported and were not verified by medical or psychological examination, although health conditions reported were medically diagnosed and parents are ideally placed to observe poor behavioral functioning and executive dysfunction in everyday life. To our knowledge, this is the largest preterm sample in a trial of DHA with a measure of behavior, and we included a range of child health problems that are common among infants born preterm.⁴⁴ A similar proportion of children in this study were receiving support at school and they had similar SDQ scores to the recent Etude Epidémiologique sur les Petits Ages Gestationnels (EPIPAGE-2) cohort of very preterm infants assessed at age 5 years.⁴⁵ The large number of comparisons increases the risk of chance findings, but we found little evidence of group differences.

Conclusions

In this follow-up of a randomized clinical trial, DHA supplementation of infants born at less than 29 weeks’ gestation did not improve behavioral functioning at age 5 years. There was no evidence of long-term adverse health outcomes with DHA supplementation.

Supplement 1.

Trial protocol and statistical analysis plan

Click here for additional data file.^{(709.1KB, pdf)}

Supplement 2.

eAppendix

eTable 1. Additional baseline characteristics of the infants and their parents

eTable 2. Additional post-randomization infant characteristics

eTable 3. Adherence and protocol deviations

eTable 4. Primary and secondary behavioral outcomes by sex

eTable 5. Secondary health outcomes at 5 years’ corrected age by sex

eTable 6. Primary and secondary behavioral outcomes by gestational age

eTable 7. Secondary health outcomes at 5 years’ corrected age by gestational age

eTable 8. Exploratory analyses of parent-rated behavior and executive functioning classified as abnormal/indicative of dysfunction at 5 years’ corrected age

Click here for additional data file.^{(335KB, pdf)}

Supplement 3.

Data sharing statement

Click here for additional data file.^{(17.6KB, pdf)}

References

1.Aarnoudse-Moens CS, Weisglas-Kuperus N, van Goudoever JB, Oosterlaan J. Meta-analysis of neurobehavioral outcomes in very preterm and/or very low birth weight children. Pediatrics. 2009;124(2):717-728. doi: 10.1542/peds.2008-2816 [DOI] [PubMed] [Google Scholar]
2.Allotey J, Zamora J, Cheong-See F, et al. Cognitive, motor, behavioural and academic performances of children born preterm: a meta-analysis and systematic review involving 64 061 children. BJOG. 2018;125(1):16-25. doi: 10.1111/1471-0528.14832 [DOI] [PubMed] [Google Scholar]
3.van Beek PE, van der Horst IE, Wetzer J, van Baar AL, Vugs B, Andriessen P. Developmental trajectories in very preterm born children up to 8 years: a longitudinal cohort study. Front Pediatr. 2021;9:672214. doi: 10.3389/fped.2021.672214 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Aarnoudse-Moens CS, Weisglas-Kuperus N, Duivenvoorden HJ, van Goudoever JB, Oosterlaan J. Executive function and IQ predict mathematical and attention problems in very preterm children. PLoS One. 2013;8(2):e55994. doi: 10.1371/journal.pone.0055994 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Johnson S, Marlow N. Early and long-term outcome of infants born extremely preterm. Arch Dis Child. 2017;102(1):97-102. doi: 10.1136/archdischild-2015-309581 [DOI] [PubMed] [Google Scholar]
6.Doyle LW, Anderson PJ. Adult outcome of extremely preterm infants. Pediatrics. 2010;126(2):342-351. doi: 10.1542/peds.2010-0710 [DOI] [PubMed] [Google Scholar]
7.Doyle LW. Are neurodevelopmental outcomes of infants born extremely preterm improving over time? Pediatrics. 2018;141(5):e20174009. doi: 10.1542/peds.2017-4009 [DOI] [PubMed] [Google Scholar]
8.Marlow N. Keeping up with outcomes for infants born at extremely low gestational ages. JAMA Pediatr. 2015;169(3):207-208. doi: 10.1001/jamapediatrics.2014.3362 [DOI] [PubMed] [Google Scholar]
9.Doyle LW, Anderson PJ. Pulmonary and neurological follow-up of extremely preterm infants. Neonatology. 2010;97(4):388-394. doi: 10.1159/000297771 [DOI] [PubMed] [Google Scholar]
10.Cheong JLY, Haikerwal A, Anderson PJ, Doyle LW. Outcomes into adulthood of infants born extremely preterm. Semin Perinatol. 2021;45(8):151483. doi: 10.1016/j.semperi.2021.151483 [DOI] [PubMed] [Google Scholar]
11.Clandinin MT, Chappell JE, Heim T, Swyer PR, Chance GW. Fatty acid utilization in perinatal de novo synthesis of tissues. Early Hum Dev. 1981;5(4):355-366. doi: 10.1016/0378-3782(81)90016-5 [DOI] [PubMed] [Google Scholar]
12.Martinez M. Tissue levels of polyunsaturated fatty acids during early human development. J Pediatr. 1992;120(4 Pt 2):S129-S138. doi: 10.1016/S0022-3476(05)81247-8 [DOI] [PubMed] [Google Scholar]
13.Twilhaar ES, Wade RM, de Kieviet JF, van Goudoever JB, van Elburg RM, Oosterlaan J. Cognitive outcomes of children born extremely or very preterm since the 1990s and associated risk factors: a meta-analysis and meta-regression. JAMA Pediatr. 2018;172(4):361-367. doi: 10.1001/jamapediatrics.2017.5323 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Brydges CR, Landes JK, Reid CL, Campbell C, French N, Anderson M. Cognitive outcomes in children and adolescents born very preterm: a meta-analysis. Dev Med Child Neurol. 2018;60(5):452-468. doi: 10.1111/dmcn.13685 [DOI] [PubMed] [Google Scholar]
15.Gould JF, Makrides M, Gibson RA, et al. Neonatal docosahexaenoic acid in preterm infants and intelligence at 5 years. N Engl J Med. 2022;387(17):1579-1588. doi: 10.1056/NEJMoa2206868 [DOI] [PubMed] [Google Scholar]
16.Collins CT, Makrides M, McPhee AJ, et al. Docosahexaenoic acid and bronchopulmonary dysplasia in preterm infants. N Engl J Med. 2017;376(13):1245-1255. doi: 10.1056/NEJMoa1611942 [DOI] [PubMed] [Google Scholar]
17.Gould JF, Roberts RM, Anderson PJ, et al. Protocol for assessing if behavioural functioning of infants born <29 weeks’ gestation is improved by omega-3 long-chain polyunsaturated fatty acids: follow-up of a randomised controlled trial. BMJ Open. 2021;11(5):e044740. doi: 10.1136/bmjopen-2020-044740 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Goodman R. The Strengths and Difficulties Questionnaire: a research note. J Child Psychol Psychiatry. 1997;38(5):581-586. doi: 10.1111/j.1469-7610.1997.tb01545.x [DOI] [PubMed] [Google Scholar]
19.Gioia GA, Espy KA, Isquith PK. The Behaviour Rating of Executive Function—Preschool Version. Psychological Assessment Resources; 2003. [Google Scholar]
20.Gioia GA, Isquith PK, Guy SC, Kenworthy L. Behavior Rating Inventory of Executive Function. 2nd ed. Psychological Assessment Resources; 2015. [Google Scholar]
21.Gould JF, Fuss BG, Roberts RM, Collins CT, Makrides M. Consequences of using chronological age versus corrected age when testing cognitive and motor development in infancy and intelligence quotient at school age for children born preterm. PLoS One. 2021;16(9):e0256824. doi: 10.1371/journal.pone.0256824 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Wilson-Ching M, Pascoe L, Doyle LW, Anderson PJ. Effects of correcting for prematurity on cognitive test scores in childhood. J Paediatr Child Health. 2014;50(3):182-188. doi: 10.1111/jpc.12475 [DOI] [PubMed] [Google Scholar]
23.Hawes DJ, Dadds MR. Australian data and psychometric properties of the Strengths and Difficulties Questionnaire. Aust N Z J Psychiatry. 2004;38(8):644-651. doi: 10.1080/j.1440-1614.2004.01427.x [DOI] [PubMed] [Google Scholar]
24.Asher MI, Keil U, Anderson HR, et al. International Study of Asthma and Allergies in Childhood (ISAAC): rationale and methods. Eur Respir J. 1995;8(3):483-491. doi: 10.1183/09031936.95.08030483 [DOI] [PubMed] [Google Scholar]
25.Varni JW, Seid M, Kurtin PS. PedsQL 4.0: reliability and validity of the Pediatric Quality of Life Inventory version 4.0 generic core scales in healthy and patient populations. Med Care. 2001;39(8):800-812. doi: 10.1097/00005650-200108000-00006 [DOI] [PubMed] [Google Scholar]
26.Rothbaum F, Weisz JR. Parental caregiving and child externalizing behavior in nonclinical samples: a meta-analysis. Psychol Bull. 1994;116(1):55-74. doi: 10.1037/0033-2909.116.1.55 [DOI] [PubMed] [Google Scholar]
27.Vernon-Feagans L, Willoughby M, Garrett-Peters P; Family Life Project Key Investigators . Predictors of behavioral regulation in kindergarten: household chaos, parenting, and early executive functions. Dev Psychol. 2016;52(3):430-441. doi: 10.1037/dev0000087 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.DeVito C, Hopkins J. Attachment, parenting, and marital dissatisfaction as predictors of disruptive behavior in preschoolers. Dev Psychopathol. 2001;13(2):215-231. doi: 10.1017/S0954579401002024 [DOI] [PubMed] [Google Scholar]
29.Epstein NB, Baldwin LM, Bishop DS. The McMaster family assessment device. J Marital Fam Ther. 1983;9:171-180. doi: 10.1111/j.1752-0606.1983.tb01497.x [DOI] [Google Scholar]
30.Mendelsohn A, Dreyer BP, Tamis-LeMonda CS. StimQ cognitive home environment questionnaire. Published 1996. Accessed February 5, 2015. https://med.nyu.edu/pediatrics/developmental/research/belle-project/stimq-cognitive-home-environment
31.Arnold DA, O’Leary SG, Wolff LS, Acker MM. The parenting scale: a measure of dysfunctional parenting in discipline situations. Psychol Assess. 1993;5(2):137-144. doi: 10.1037/1040-3590.5.2.137 [DOI] [Google Scholar]
32.Sullivan TR, White IR, Salter AB, Ryan P, Lee KJ. Should multiple imputation be the method of choice for handling missing data in randomized trials? Stat Methods Med Res. 2018;27(9):2610-2626. doi: 10.1177/0962280216683570 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.White IR, Royston P, Wood AM. Multiple imputation using chained equations: Issues and guidance for practice. Stat Med. 2011;30(4):377-399. doi: 10.1002/sim.4067 [DOI] [PubMed] [Google Scholar]
34.Hewawasam E, Collins CT, Muhlhausler BS, et al. DHA supplementation in infants born preterm and the effect on attention at 18 months’ corrected age: follow-up of a subset of the N3RO randomised controlled trial. Br J Nutr. 2021;125(4):420-431. doi: 10.1017/S0007114520002500 [DOI] [PubMed] [Google Scholar]
35.Guillot M, Synnes A, Pronovost E, et al. Maternal high-dose dha supplementation and neurodevelopment at 18-22 months of preterm children. Pediatrics. 2022;150(1):e2021055819. doi: 10.1542/peds.2021-055819 [DOI] [PubMed] [Google Scholar]
36.Collins CT, Gibson RA, Anderson PJ, et al. Neurodevelopmental outcomes at 7 years’ corrected age in preterm infants who were fed high-dose docosahexaenoic acid to term equivalent: a follow-up of a randomised controlled trial. BMJ Open. 2015;5(3):e007314. doi: 10.1136/bmjopen-2014-007314 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Smithers LG, Gibson RA, McPhee A, Makrides M. Higher dose of docosahexaenoic acid in the neonatal period improves visual acuity of preterm infants: results of a randomized controlled trial. Am J Clin Nutr. 2008;88(4):1049-1056. doi: 10.1093/ajcn/88.4.1049 [DOI] [PubMed] [Google Scholar]
38.Gould JF, Best K, Netting MJ, Gibson RA, Makrides M. New methodologies for conducting maternal, infant, and child nutrition research in the era of COVID-19. Nutrients. 2021;13(3):941. doi: 10.3390/nu13030941 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Gould JF, Smithers LG. Prenatal n-3 long-chain polyunsaturated fatty acids and children’s executive functions. In: Watson RR, Preedy VR, eds. Omega Fatty Acids in Brain and Neurological Health (Second Edition). Academic Press; 2019:83-105. [Google Scholar]
40.Sullivan TR, Gould JF, Bednarz JM, et al. Mediation analysis to untangle opposing associations of high-dose docosahexaenoic acid with IQ and bronchopulmonary dysplasia in children born preterm. JAMA Netw Open. 2023;6(6):e2317870. doi: 10.1001/jamanetworkopen.2023.17870 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Best KP, Sullivan T, Palmer D, et al. Prenatal fish oil supplementation and allergy: 6-year follow-up of a randomized controlled trial. Pediatrics. 2016;137(6):e20154443. doi: 10.1542/peds.2015-4443 [DOI] [PubMed] [Google Scholar]
42.Lapillonne A, Eleni dit Trolli S, Kermorvant-Duchemin E. Postnatal docosahexaenoic acid deficiency is an inevitable consequence of current recommendations and practice in preterm infants. Neonatology. 2010;98(4):397-403. doi: 10.1159/000320159 [DOI] [PubMed] [Google Scholar]
43.Blencowe H, Lee AC, Cousens S, et al. Preterm birth-associated neurodevelopmental impairment estimates at regional and global levels for 2010. Pediatr Res. 2013;74(Suppl 1)(suppl 1):17-34. doi: 10.1038/pr.2013.204 [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Doyle LW, Anderson PJ, Battin M, et al. Long term follow up of high risk children: who, why and how? BMC Pediatr. 2014;14:279. doi: 10.1186/1471-2431-14-279 [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Pierrat V, Marchand-Martin L, Marret S, et al. ; EPIPAGE-2 writing group . Neurodevelopmental outcomes at age 5 among children born preterm: EPIPAGE-2 cohort study. BMJ. 2021;373(741):n741. doi: 10.1136/bmj.n741 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1.

Trial protocol and statistical analysis plan

Click here for additional data file.^{(709.1KB, pdf)}

Supplement 2.

eAppendix

eTable 1. Additional baseline characteristics of the infants and their parents

eTable 2. Additional post-randomization infant characteristics

eTable 3. Adherence and protocol deviations

eTable 4. Primary and secondary behavioral outcomes by sex

eTable 5. Secondary health outcomes at 5 years’ corrected age by sex

eTable 6. Primary and secondary behavioral outcomes by gestational age

eTable 7. Secondary health outcomes at 5 years’ corrected age by gestational age

eTable 8. Exploratory analyses of parent-rated behavior and executive functioning classified as abnormal/indicative of dysfunction at 5 years’ corrected age

Click here for additional data file.^{(335KB, pdf)}

Supplement 3.

Data sharing statement

Click here for additional data file.^{(17.6KB, pdf)}

[poi230076r1] 1.Aarnoudse-Moens CS, Weisglas-Kuperus N, van Goudoever JB, Oosterlaan J. Meta-analysis of neurobehavioral outcomes in very preterm and/or very low birth weight children. Pediatrics. 2009;124(2):717-728. doi: 10.1542/peds.2008-2816 [DOI] [PubMed] [Google Scholar]

[poi230076r2] 2.Allotey J, Zamora J, Cheong-See F, et al. Cognitive, motor, behavioural and academic performances of children born preterm: a meta-analysis and systematic review involving 64 061 children. BJOG. 2018;125(1):16-25. doi: 10.1111/1471-0528.14832 [DOI] [PubMed] [Google Scholar]

[poi230076r3] 3.van Beek PE, van der Horst IE, Wetzer J, van Baar AL, Vugs B, Andriessen P. Developmental trajectories in very preterm born children up to 8 years: a longitudinal cohort study. Front Pediatr. 2021;9:672214. doi: 10.3389/fped.2021.672214 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi230076r4] 4.Aarnoudse-Moens CS, Weisglas-Kuperus N, Duivenvoorden HJ, van Goudoever JB, Oosterlaan J. Executive function and IQ predict mathematical and attention problems in very preterm children. PLoS One. 2013;8(2):e55994. doi: 10.1371/journal.pone.0055994 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi230076r5] 5.Johnson S, Marlow N. Early and long-term outcome of infants born extremely preterm. Arch Dis Child. 2017;102(1):97-102. doi: 10.1136/archdischild-2015-309581 [DOI] [PubMed] [Google Scholar]

[poi230076r6] 6.Doyle LW, Anderson PJ. Adult outcome of extremely preterm infants. Pediatrics. 2010;126(2):342-351. doi: 10.1542/peds.2010-0710 [DOI] [PubMed] [Google Scholar]

[poi230076r7] 7.Doyle LW. Are neurodevelopmental outcomes of infants born extremely preterm improving over time? Pediatrics. 2018;141(5):e20174009. doi: 10.1542/peds.2017-4009 [DOI] [PubMed] [Google Scholar]

[poi230076r8] 8.Marlow N. Keeping up with outcomes for infants born at extremely low gestational ages. JAMA Pediatr. 2015;169(3):207-208. doi: 10.1001/jamapediatrics.2014.3362 [DOI] [PubMed] [Google Scholar]

[poi230076r9] 9.Doyle LW, Anderson PJ. Pulmonary and neurological follow-up of extremely preterm infants. Neonatology. 2010;97(4):388-394. doi: 10.1159/000297771 [DOI] [PubMed] [Google Scholar]

[poi230076r10] 10.Cheong JLY, Haikerwal A, Anderson PJ, Doyle LW. Outcomes into adulthood of infants born extremely preterm. Semin Perinatol. 2021;45(8):151483. doi: 10.1016/j.semperi.2021.151483 [DOI] [PubMed] [Google Scholar]

[poi230076r11] 11.Clandinin MT, Chappell JE, Heim T, Swyer PR, Chance GW. Fatty acid utilization in perinatal de novo synthesis of tissues. Early Hum Dev. 1981;5(4):355-366. doi: 10.1016/0378-3782(81)90016-5 [DOI] [PubMed] [Google Scholar]

[poi230076r12] 12.Martinez M. Tissue levels of polyunsaturated fatty acids during early human development. J Pediatr. 1992;120(4 Pt 2):S129-S138. doi: 10.1016/S0022-3476(05)81247-8 [DOI] [PubMed] [Google Scholar]

[poi230076r13] 13.Twilhaar ES, Wade RM, de Kieviet JF, van Goudoever JB, van Elburg RM, Oosterlaan J. Cognitive outcomes of children born extremely or very preterm since the 1990s and associated risk factors: a meta-analysis and meta-regression. JAMA Pediatr. 2018;172(4):361-367. doi: 10.1001/jamapediatrics.2017.5323 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi230076r14] 14.Brydges CR, Landes JK, Reid CL, Campbell C, French N, Anderson M. Cognitive outcomes in children and adolescents born very preterm: a meta-analysis. Dev Med Child Neurol. 2018;60(5):452-468. doi: 10.1111/dmcn.13685 [DOI] [PubMed] [Google Scholar]

[poi230076r15] 15.Gould JF, Makrides M, Gibson RA, et al. Neonatal docosahexaenoic acid in preterm infants and intelligence at 5 years. N Engl J Med. 2022;387(17):1579-1588. doi: 10.1056/NEJMoa2206868 [DOI] [PubMed] [Google Scholar]

[poi230076r16] 16.Collins CT, Makrides M, McPhee AJ, et al. Docosahexaenoic acid and bronchopulmonary dysplasia in preterm infants. N Engl J Med. 2017;376(13):1245-1255. doi: 10.1056/NEJMoa1611942 [DOI] [PubMed] [Google Scholar]

[poi230076r17] 17.Gould JF, Roberts RM, Anderson PJ, et al. Protocol for assessing if behavioural functioning of infants born <29 weeks’ gestation is improved by omega-3 long-chain polyunsaturated fatty acids: follow-up of a randomised controlled trial. BMJ Open. 2021;11(5):e044740. doi: 10.1136/bmjopen-2020-044740 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi230076r18] 18.Goodman R. The Strengths and Difficulties Questionnaire: a research note. J Child Psychol Psychiatry. 1997;38(5):581-586. doi: 10.1111/j.1469-7610.1997.tb01545.x [DOI] [PubMed] [Google Scholar]

[poi230076r19] 19.Gioia GA, Espy KA, Isquith PK. The Behaviour Rating of Executive Function—Preschool Version. Psychological Assessment Resources; 2003. [Google Scholar]

[poi230076r20] 20.Gioia GA, Isquith PK, Guy SC, Kenworthy L. Behavior Rating Inventory of Executive Function. 2nd ed. Psychological Assessment Resources; 2015. [Google Scholar]

[poi230076r21] 21.Gould JF, Fuss BG, Roberts RM, Collins CT, Makrides M. Consequences of using chronological age versus corrected age when testing cognitive and motor development in infancy and intelligence quotient at school age for children born preterm. PLoS One. 2021;16(9):e0256824. doi: 10.1371/journal.pone.0256824 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi230076r22] 22.Wilson-Ching M, Pascoe L, Doyle LW, Anderson PJ. Effects of correcting for prematurity on cognitive test scores in childhood. J Paediatr Child Health. 2014;50(3):182-188. doi: 10.1111/jpc.12475 [DOI] [PubMed] [Google Scholar]

[poi230076r23] 23.Hawes DJ, Dadds MR. Australian data and psychometric properties of the Strengths and Difficulties Questionnaire. Aust N Z J Psychiatry. 2004;38(8):644-651. doi: 10.1080/j.1440-1614.2004.01427.x [DOI] [PubMed] [Google Scholar]

[poi230076r24] 24.Asher MI, Keil U, Anderson HR, et al. International Study of Asthma and Allergies in Childhood (ISAAC): rationale and methods. Eur Respir J. 1995;8(3):483-491. doi: 10.1183/09031936.95.08030483 [DOI] [PubMed] [Google Scholar]

[poi230076r25] 25.Varni JW, Seid M, Kurtin PS. PedsQL 4.0: reliability and validity of the Pediatric Quality of Life Inventory version 4.0 generic core scales in healthy and patient populations. Med Care. 2001;39(8):800-812. doi: 10.1097/00005650-200108000-00006 [DOI] [PubMed] [Google Scholar]

[poi230076r26] 26.Rothbaum F, Weisz JR. Parental caregiving and child externalizing behavior in nonclinical samples: a meta-analysis. Psychol Bull. 1994;116(1):55-74. doi: 10.1037/0033-2909.116.1.55 [DOI] [PubMed] [Google Scholar]

[poi230076r27] 27.Vernon-Feagans L, Willoughby M, Garrett-Peters P; Family Life Project Key Investigators . Predictors of behavioral regulation in kindergarten: household chaos, parenting, and early executive functions. Dev Psychol. 2016;52(3):430-441. doi: 10.1037/dev0000087 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi230076r28] 28.DeVito C, Hopkins J. Attachment, parenting, and marital dissatisfaction as predictors of disruptive behavior in preschoolers. Dev Psychopathol. 2001;13(2):215-231. doi: 10.1017/S0954579401002024 [DOI] [PubMed] [Google Scholar]

[poi230076r29] 29.Epstein NB, Baldwin LM, Bishop DS. The McMaster family assessment device. J Marital Fam Ther. 1983;9:171-180. doi: 10.1111/j.1752-0606.1983.tb01497.x [DOI] [Google Scholar]

[poi230076r30] 30.Mendelsohn A, Dreyer BP, Tamis-LeMonda CS. StimQ cognitive home environment questionnaire. Published 1996. Accessed February 5, 2015. https://med.nyu.edu/pediatrics/developmental/research/belle-project/stimq-cognitive-home-environment

[poi230076r31] 31.Arnold DA, O’Leary SG, Wolff LS, Acker MM. The parenting scale: a measure of dysfunctional parenting in discipline situations. Psychol Assess. 1993;5(2):137-144. doi: 10.1037/1040-3590.5.2.137 [DOI] [Google Scholar]

[poi230076r32] 32.Sullivan TR, White IR, Salter AB, Ryan P, Lee KJ. Should multiple imputation be the method of choice for handling missing data in randomized trials? Stat Methods Med Res. 2018;27(9):2610-2626. doi: 10.1177/0962280216683570 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi230076r33] 33.White IR, Royston P, Wood AM. Multiple imputation using chained equations: Issues and guidance for practice. Stat Med. 2011;30(4):377-399. doi: 10.1002/sim.4067 [DOI] [PubMed] [Google Scholar]

[poi230076r34] 34.Hewawasam E, Collins CT, Muhlhausler BS, et al. DHA supplementation in infants born preterm and the effect on attention at 18 months’ corrected age: follow-up of a subset of the N3RO randomised controlled trial. Br J Nutr. 2021;125(4):420-431. doi: 10.1017/S0007114520002500 [DOI] [PubMed] [Google Scholar]

[poi230076r35] 35.Guillot M, Synnes A, Pronovost E, et al. Maternal high-dose dha supplementation and neurodevelopment at 18-22 months of preterm children. Pediatrics. 2022;150(1):e2021055819. doi: 10.1542/peds.2021-055819 [DOI] [PubMed] [Google Scholar]

[poi230076r36] 36.Collins CT, Gibson RA, Anderson PJ, et al. Neurodevelopmental outcomes at 7 years’ corrected age in preterm infants who were fed high-dose docosahexaenoic acid to term equivalent: a follow-up of a randomised controlled trial. BMJ Open. 2015;5(3):e007314. doi: 10.1136/bmjopen-2014-007314 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi230076r37] 37.Smithers LG, Gibson RA, McPhee A, Makrides M. Higher dose of docosahexaenoic acid in the neonatal period improves visual acuity of preterm infants: results of a randomized controlled trial. Am J Clin Nutr. 2008;88(4):1049-1056. doi: 10.1093/ajcn/88.4.1049 [DOI] [PubMed] [Google Scholar]

[poi230076r38] 38.Gould JF, Best K, Netting MJ, Gibson RA, Makrides M. New methodologies for conducting maternal, infant, and child nutrition research in the era of COVID-19. Nutrients. 2021;13(3):941. doi: 10.3390/nu13030941 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi230076r39] 39.Gould JF, Smithers LG. Prenatal n-3 long-chain polyunsaturated fatty acids and children’s executive functions. In: Watson RR, Preedy VR, eds. Omega Fatty Acids in Brain and Neurological Health (Second Edition). Academic Press; 2019:83-105. [Google Scholar]

[poi230076r40] 40.Sullivan TR, Gould JF, Bednarz JM, et al. Mediation analysis to untangle opposing associations of high-dose docosahexaenoic acid with IQ and bronchopulmonary dysplasia in children born preterm. JAMA Netw Open. 2023;6(6):e2317870. doi: 10.1001/jamanetworkopen.2023.17870 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi230076r41] 41.Best KP, Sullivan T, Palmer D, et al. Prenatal fish oil supplementation and allergy: 6-year follow-up of a randomized controlled trial. Pediatrics. 2016;137(6):e20154443. doi: 10.1542/peds.2015-4443 [DOI] [PubMed] [Google Scholar]

[poi230076r42] 42.Lapillonne A, Eleni dit Trolli S, Kermorvant-Duchemin E. Postnatal docosahexaenoic acid deficiency is an inevitable consequence of current recommendations and practice in preterm infants. Neonatology. 2010;98(4):397-403. doi: 10.1159/000320159 [DOI] [PubMed] [Google Scholar]

[poi230076r43] 43.Blencowe H, Lee AC, Cousens S, et al. Preterm birth-associated neurodevelopmental impairment estimates at regional and global levels for 2010. Pediatr Res. 2013;74(Suppl 1)(suppl 1):17-34. doi: 10.1038/pr.2013.204 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi230076r44] 44.Doyle LW, Anderson PJ, Battin M, et al. Long term follow up of high risk children: who, why and how? BMC Pediatr. 2014;14:279. doi: 10.1186/1471-2431-14-279 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi230076r45] 45.Pierrat V, Marchand-Martin L, Marret S, et al. ; EPIPAGE-2 writing group . Neurodevelopmental outcomes at age 5 among children born preterm: EPIPAGE-2 cohort study. BMJ. 2021;373(741):n741. doi: 10.1136/bmj.n741 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

High-Dose Docosahexaenoic Acid in Newborns Born at Less Than 29 Weeks’ Gestation and Behavior at Age 5 Years

Jacqueline F Gould, PhD

Rachel M Roberts, PhD

Peter J Anderson, PhD

Maria Makrides, PhD

Thomas R Sullivan, PhD

Robert A Gibson, PhD

Andrew J McPhee, MB, BS

Lex W Doyle, MD

Jana M Bednarz, GDip

Karen P Best, PhD

Gillian Opie, MB, BS

Javeed Travadi, DM

Jeanie L Y Cheong, MD

Peter G Davis, MD

Mary Sharp, MB, BS

Karen Simmer, PhD

Kenneth Tan, PhD

Scott Morris, PhD

Kei Lui, MD

Srinivas Bolisetty, MD

Helen Liley, MB, ChB

Jacqueline Stack, MB, ChB

Carmel T Collins, PhD

Key Points

Question

Findings

Meaning

Abstract

Importance

Objective

Design, Setting and Participants

Interventions

Main Outcomes and Measures

Results

Conclusions and Relevance

Trial Registration

Introduction

Methods

Study Design

Participants

Randomization and Blinding

Intervention

Outcomes and Measures

Sample Characteristics

Statistical Analysis

Results

Figure. Participant Flow Diagram.

Table 1. Baseline and Postrandomization Characteristics of the Children and Their Parents or Caregivers.

Primary and Secondary Outcomes

Table 2. Primary Outcome and Secondary Behavioral Outcomes at 5 Years’ Corrected Age.

Table 3. Secondary Health Outcomes at 5 Years’ Corrected Age.

Subgroup and Exploratory Analyses

Discussion

Strengths and Limitations

Conclusions

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases