Effect of long-chain polyunsaturated fatty acids in infant formula on long-term cognitive function in childhood: A systematic review and meta-analysis of randomised controlled trials.

STUDY REGISTRATION: PROSPERO registration numbers CRD42018105196 and CRD42018088868.

Introduction

Long-chain polyunsaturated fatty acids (LCPUFA), such as docosahexaenoic acid (DHA) and arachidonic acid (AA), are important structural components of the human brain that mainly accumulate during the third trimester of pregnancy and early infancy [1-3]. Human breast milk contains DHA, AA, and their fatty acid precursors [4] but, historically, infant formula contained only the precursors alpha linoleic acid and linoleic acid, which infants, especially those born preterm, may not be able to effectively synthesise into DHA and AA [5].

Research suggests that breast-fed children have higher cognitive ability compared to formula-fed children [6-9]. Lack of preformed LCPUFA in infant formula has been hypothesised as contributing to these cognitive differences. Yet so far there is no clear evidence from published randomised controlled trials (RCTs) that LCPUFA-supplemented infant formula improves cognition compared with unsupplemented formula milk [5, 10]. Previous systematic reviews of RCTs may have failed to detect a difference in cognition because they mainly focused on early measures of cognition, such as Bayley Scales of Infant Development. Early measures of cognition are, however, not adequate to differentiate between cognitive skills potentially affected by nutrient supplementation and are poorly predictive of cognition during school age [11-13]. Follow-up later in childhood, using more reliable measures of cognitive function such as Intelligence Quotient (IQ) scores, might be more likely to detect an existing effect of LCPUFA-supplementation.

Clear evidence on the long-term effects of LCPUFA-supplementation is needed as the EU Commission (EC) recently mandated the addition of one type of LCPUFA, DHA, to all infant and follow-on formulae [14]. While the decision acknowledged the lack of evidence on cognitive benefits and was instead based around theoretical arguments, supplementation comes at a cost: a family can spend up to $400 extra per year on LCPUFA-supplemented- compared to unsupplemented infant formulas and mandatory supplementation may result in price rises across the market [15].

To our knowledge, no systematic review has previously focused on later childhood -when more accurate measures are available [11-13]- to study the cognitive effects of infant formula supplemented with LCPUFA. ‏The present study combines published and previously unpublished trial data, acquired through contacting trial authors, to compare the long-term cognitive effects of LCPUFA-supplemented versus unsupplemented infant formula in children born at term and preterm.

Methods

Search strategy

This systematic review and meta-analysis follows two published protocols (one for terms and one for preterms) [16, 17], based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [18]. We searched Medline, Embase, proceedings from major scientific meetings of child nutrition (S2 File) and the Cochrane Central Register of Controlled Trials in October 2019, without date or language restrictions. We reviewed the reference lists of the included studies and traced subsequent publications. We first identified RCT participant cohorts based on any infant formula supplementation with LCPUFA, independent of whether cognitive outcomes were reported. We then contacted a total of 18 trialists, ethics committees or industry representatives to identify potential unpublished data, clarify study details and to ask whether they knew of any other eligible trials that had measured cognitive outcomes ≥2.5 years of age (S4 and S5 Tables in S1 File). The cut-off of ≥2.5 years was based on the age where early development of the prefrontal cortex, the brain region associated with higher cognitive functions is completed [19]. Abstract review was done in duplicate by MV and SD; consensus was achieved by discussion. All extracted data is available within S1 Table in S1 File.

Inclusion and exclusion criteria

We based our selection on trial cohorts rather than publications. We included trial cohorts where infants were given either infant formula supplemented with LCPUFA (DHA alone or DHA together with AA, at any dose) compared with unsupplemented formula. Trial cohorts were eligible if commencement of the intervention began within 2 weeks of birth and they measured cognitive function age ≥ 2.5 years using validated measures including Wechsler and Stanford-Binet IQ scores. We excluded trial cohorts for which we could not find any cognitive outcomes ≥2.5 years of age (irrespective of whether this outcome was published).

Outcomes and data analysis

The primary outcome was the pooled difference of cognitive ability between supplemented and unsupplemented groups. We decided to use the cognitive test reported most frequently among the included studies rather than a combination of different tests, to increase the interpretability of the primary outcome and decrease heterogeneity. We aimed to use the mean difference (MD) when the cognitive measure was already standardised (e.g. IQ score) so as to increase interpretability, otherwise we would use the standardised mean difference (SMD). We performed separate analyses for term and preterm-born participants because healthy term infants are able to synthesise LCPUFA from fatty acid pre-cursors, whereas preterm babies are born with fewer LCPUFA reserves accumulated in utero and are less able to synthesise LCPUFA than term-born babies. We therefore hypothesised that term/preterm status could modify the effect of supplemented LCPUFA.

The secondary outcome was the pooled SMD of all available cognitive test scores. To not include the same participants multiple times, we only included one score per trial cohort. For this, we used the score at the oldest age available under the assumption that scores at a later age are a more accurate reflection of cognitive ability than scores at an earlier age.

All analyses were performed in RevMan v5.4 and included participants with the relevant outcome in the groups to which they were randomised. We defined statistically significant differences based on a p-value <0.05 and report all summary measures along with 95% confidence intervals and a measure of heterogeneity (I). We also calculated the prediction interval (PI) to accurately reflect any uncertainty about clinical harms and benefits [20].

Strength of evidence and risk of bias assessment

We assessed the strength of evidence and risk of bias for each study using GRADE (S3 Table in S1 File) and the Cochrane Risk of Bias Tool II, which was also used as the template for data extraction (S2 Table in S1 File). Post-hoc, we explored potential for publication bias by plotting the SMDs for all available scores against their standard error (SE), to visualise the relative distribution of published and unpublished outcomes (S1 Fig in S1 File).

Results

We included eight unique trial cohorts [21-49] of which six were performed in infants born at term and two in infants born preterm (Fig 1). We obtained previously unpublished outcome data for two RCTs: Firstly, a two-centre trial of term babies in England [30, 31] provided unpublished follow-up data on IQ assessments using the Wechsler Preschool and Primary Scales of Intelligence (WPPSI) at age 4.5 years and the Wechsler Abbreviated Scales of Intelligence (WASI) at age 16 years. Secondly, a two-centre trial in England of babies born preterm provided unpublished data from their IQ assessments using the WASI at age 16 years [50]. We also received partly published outcome data from three trials, in a form that allowed them to be included in a pooled meta-analysis: IQ using the WASI at 9 years from a Dutch trial of term babies [35-38]; IQ using the WASI at age 16 years from the Kansas centre of the US based DIAMOND trial conducted in term infants [39, 41–46, 51]; and IQ assessments adjusted for maternal education at 10 years from a two-centre study in Scottish preterm children [48, 49].

Fig 1

Study selection process.

Table 1 shows the characteristics of all included trial cohorts. The total number of children randomised was not reported for one RCT [24-29]. Study randomisation was performed between 1992 and 2004 with the latest cognitive assessments conducted at mean ages 3.3–16 years. All studies randomised participants to infant formula supplemented with LCPUFA containing DHA and AA or to unsupplemented infant formula. DHA was sourced from egg, fish, fungi, algae or starflower oil and made up between 0.12 and 0.96% of total fat content. Ratios of DHA:AA ranged from 1:0.8 to 1:3.6. Duration of the intervention ranged from two to 12 months in term infants and three weeks to 9 months in infants born preterm. All trials, except for one, included a non-randomised breast-fed reference group but these were not analysed in this review. Randomised children in one preterm trial [48, 49] could receive some breastmilk during the first months, but the intake was balanced across groups.

Table 1

Characteristics of included RCT cohorts and associated publications.

RCT Cohort (recruitment years) and [publication references]		Criteria	LCPUFA, %	Any breastmilk during study	Breastfed reference group	Start LCPUFA (age in days)	Duration LCPUFA (months)	Outcomes (age in years)	Unpublished outcome data?	%Follow-up: n followed-up1,2/ n randomised2
Term studies		wga
US: 3 centres (92–93) [21–23]		>36	Egg DHA 0.12 + Egg AA 0.43	No	Yes	<7 d	12	PPVT (3.3) SB IQ (3.3)		79%	72/91
Europe: 6 centres* (92–93) [24–29]		37–42	Egg DHA 0.30§+ Egg AA 0.44§	No	Yes	<3 d	4	WPPSI-R (6)		n/a	147/ n/a**
ENG: 2 centres (93–95) [30, 31]		>36	Egg DHA 0.32 + Egg AA 0.30	No	Yes	<7 d	6	WPPSI-R (4.5) WASI (16)	Yes	60%	184/309
US: Dallas (93–95) [32–34]		37–40	Algae DHA 0.36 + Fungi AA 0.72	No	Yes	<5 d	3.9	WPPSI-R (4)		68%	36/53
NL: Groningen (97–99) [35–38]		37–42	Egg & Fish DHA 0.30 + Fungi AA 0.45	No	Yes	<5 d	2	WASI (9)	partly‡	68%	214/314
US: DIAMOND [39–46]	Dallas (02–04)	37–42	Algae DHA 0.32 + Fungi AA 0.64	No	No	<10 d	4	PPVT (3.5), BBCS-R (2.5)		46%	42/92
	Kansas (02–04)			No				WPPSI-R (6)	partly‡	38%	30/80
Preterm studies		bw, wga
ENG: 2 centres (93–96) [47]		<1750g, <37	Algae DHA 0.32 + Fungi AA 0.64	No	Yes	<11 d	0.69	WASI (16)	yes	9%	17/196
SCT: Glasgow (95–97) [48, 49]		≤2000g, <35	Egg DHA 0.17 + Egg AA 0.31	Yes	No	2–60 d	9	WASI (10)	partly‡	45%	107/238

bw: birthweight, wga: Weeks gestational age, PPVT: Peabody Picture Vocabulary Test, SB IQ Stanford-Binet IQ, BBCS-R: Bracken Basic Concept Scale-Revised, WASI: Wechsler Abbreviated Scale of Intelligence, WPPSI-R: Wechsler Preschool and Primary Scale of Intelligence-Revised;

*Two locations unknown, 1: When latest cognitive outcome was measured 2: Only dose/source of interest, some studies had more than one randomised dose/source group (only one per trial is included here);

§Two different concentrations were published: DHA = 0.30 [29] or 0.21 [54]–AA = 0.44 [29] or 0.35 [54])

‡data was published in graphical form or differently modelled; n/a: Not available;

** the 4 centres that were followed up randomised 237 infants between them but it is unknown how many were randomised in the remaining two centres and why they weren’t followed-up. It follows that the follow-up rate was < 62%.

Three RCTs had more than one randomised intervention group [21, 22, 32–34, 39–46, 52]. To ensure comparability with a previous Cochrane review [53], we included only the intervention group in our analysis that was most similar in DHA dose and source to the other included RCTs. One study [39-Early Hum Dev. 2013 ">46] randomised babies in two centres and then conducted different cognitive assessments at different ages for children followed-up stratified by centre. We regarded these as independent and included both reported cognitive assessments in our analysis.

The most frequently reported measures of cognitive function were the WPPSI IQ at ages 4–6 years (in four term RCTs) and the WASI IQ at ages 9–16 years (two term and two preterm RCTs). Among term born participants, assessments at the oldest age comprised WASI at ages 9 and 16 years (2 RCTs), Stanford-Binet IQ at age 3.3 years (1 RCT), the Peabody Picture Vocabulary Test at age 3.5 years (1 RCT) and WPPSI 4–6 years (3 RCTs; see Table 1). Other reported measures in term were the Bracken Basic Concept Scale-Revised at age 2.5 years (one term RCT). Among preterm-born participants, the most frequently reported assessments were also those conducted at the oldest age.

Primary outcome

We pooled the data for infants born at term using random-effects because heterogeneity was judged to be high with an I of 72% (p = 0.01), despite homogeneity in terms of the assessment used. Fig 2 shows that, among term-born babies, the pooled MD from four trials suggests no difference at the 5% level in the WPPSI score between LCPUFA supplemented and control groups: MD -0.04 IQ points. Uncertainty around the effect estimate was extremely high: 95% CI -5.94 to 5.85 and 95% PI from -14.17 to 14.25.

Fig 2

Primary outcome term infants: WPPSI-R IQ at ages 4–6 years (mean difference).

We pooled the data for infants born preterm also using random-effects because heterogeneity was judged to be high with an I of 83% (p = 0.01), despite homogeneity in terms of the assessment used. Fig 3 shows that among preterm-born babies, the pooled MD suggests no difference in the WASI Scale IQ at age 9–16 years: MD -7.71 IQ points. Again, uncertainty around the effect estimate was extremely high: 95% CI -24.63 to 9.22 and 95% PI from -97.80 to 82.38.

Fig 3

Primary outcome preterm infants: WASI IQ at ages 10–17 years (mean difference).

Secondary outcomes

Pooled cognitive tests scores performed at the oldest available age in each trial showed no evidence that children who received LCPUFA-supplemented infant formula differed from the control group: the SMD for term born children using random effects was -0.10, with a 95% CI of -0.32 to 0.12, with a 95% PI of -0.61 to 0.39 (Fig 4). The cognitive measures included were the Peabody Picture Vocabulary Test (PPVT) at age 3.5 years, Stanford-Binet IQ at 3.3 years, WASI at 9 and 16 years, and the WPPSI at age 4 and 6 years. There were no further available cognitive measures for infants born preterm.

Fig 4

Secondary outcome term infants: Cognitive function summary score (standardised mean difference).

Fig 5 displays the results from the Cochrane Risk of Bias Tool. It highlights that potential for bias from attrition of study participants was a universal problem for all included trials. This is also reflected in the GRADE Summary of Findings (S3 Table in S1 File). Overall, the quality of available evidence was low, rated down for heterogeneity, attrition and potential for bias from selective publication. Potential for selective publication was based on correspondence from trialists about the (perceived) difficulty of publishing harmful results. S1 Fig in S1 File plots published and unpublished effect estimates against their standard error and indicates that unpublished effect estimates tend to be those that show harm. However, this should be interpreted with caution since visual analysis of potential for publication bias has limited reliability with <10 studies [55]. Completeness of follow-up for cognitive assessment was low, ranging from 9% to 79% of children initially randomised (median 52.6). Most studies reported balanced group characteristics at follow-up (S2 Table in S1 File).

Fig 5

Risk of bias summary: Based on latest available included cognitive outcome.

Discussion

We found no evidence that LCPUFA-supplemented infant formula improved long-term cognition among children born at term or preterm. Effect estimates were highly uncertain and included potential for large benefit and large harm.

This uncertainty should be taken seriously. While previous trials on LCPUFA-supplementation mostly reported either no effect or transiently favourable effects on developmental outcomes, negative effects are not without precedent. LCPUFA-supplementation has been associated with adverse effects on growth (including head growth) [56-58] and on development such as reduced vocabulary scores in term infants at age 14 months [52]. Furthermore, LCPUFA formula supplementation comprising DHA has been associated with potential harms in other domains, for example a higher risk of bronchopulmonary dysplasia in preterm infants [59].

Potential harms of LCPUFA might relate to the source of LCPUFA. LCPUFA in the included studies of this review were derived from egg, fish, algae and fungi and may not have the same functional effects as LCPUFA present in human milk. Furthermore, the DHA content of human milk is variable and heavily influenced by maternal diet. It therefore does not easily translate into an optimal dose [60]. This incertitude was reflected in the variety of doses administered in the included trials and likely drove the substantial heterogeneity that was observed in our meta-analyses. The plotted SMDs of all available outcomes also suggest that heterogeneity between studies is partly due to under-representation of outcomes showing a harmful effect of LCPUFA. We cannot determine the reason for non-publication for all trials, but a (perceived) difficulty in publishing negative results might play a role. Apart from that, trials were conducted in similar populations, using similar inclusion criteria, and the pooled primary outcomes were based on the same respective test.

We were not able to perform subgroup analyses to determine which factors accounted for the observed heterogeneity due to the small number of available trials. Yet even if findings are truly inconsistent, this would not change the conclusion of this study, namely that current evidence for supplementing infant formula with LCPUFA on the basis of cognitive benefits is weak and does not exclude potential for large harm.

Strengths of this review include a comprehensive search independent of reported outcomes, duplicate assessment of eligibility, and rigorous application of the GRADE and Cochrane Risk of Bias approach to rate quality of evidence and risk of bias. In contrast with previous systematic reviews, we included previously unpublished outcomes, which enabled a meta-analysis of measures of cognitive function beyond the first two years of life. As the child ages, cognitive assessments become more discriminatory and predictive of adult function than tests before two years of age and less dependent on the situation and outcome assessor [11-13].

The limitations of our review are related to the quality of the underlying evidence. We observed substantial statistical heterogeneity, potential for attrition bias in all outcomes, as well as potential for selective publication. Attrition of study participants is a universal problem in long-term nutrition studies [56-58]. Potentially negative effects of LCPUFA-supplementation could have resulted from selective follow-up. However, it seems unlikely that either 1) children with lower IQ who were previously assigned to the control group would be less likely to respond than children with lower IQ from the intervention group, or 2) children with higher IQ outcomes be more likely to respond if they were in the control group, especially as all trials were blinded.

While there was industry involvement in all of the original trials, only four of the ten follow-up studies reported in our analyses received industry funding. In six of the eight trials there were other potential conflicts of interest (S7 Table in S1 File). While industry involvement is very common in the area of infant nutrition studies it is not necessarily predictive of lower study quality. It is also necessary to emphasise that many of the outcomes included in our review were published at a time when reporting standards were lower than today. Importantly, the included studies represent the only available evidence on long-term cognitive outcomes.

More robust evidence of benefit, and certainty about absence of harm, is needed to justify mandatory LCPUFA-supplementation of infant formula. New trials would take time, are expensive and would suffer from the same problems of attrition and resulting biases as the studies included in this review. Recent methods, involving data linkage of extant trial data to administrative education and health data in adolescence and adulthood, offer a more rapid, less biased, and cost effective way of obtaining data on long-term outcomes. Linkage of historical trials to administrative education or health datasets is achievable where trial data and participant identifiers have been retained and governance arrangements allow secure linkage without re-consent [61, 62].

We found no evidence that LCPUFA-supplemented infant formula benefits cognitive function compared with unsupplemented formula in children born at term or preterm. Given the lack of benefit on other functional outcomes [5, 53] and the additional costs of supplemented formula [63], widespread addition of LCPUFA to infant and follow-on formula cannot be supported until further robust evidence excludes potential for future harm.

(DOCX)

Click here for additional data file.

(DOC)

Click here for additional data file.

13 Aug 2020

PONE-D-20-14949

Effect of long-chain polyunsaturated fatty acids in infant formula on long-term ‎cognitive function in childhood: a systematic review and meta-analysis of randomised controlled trials

PLOS ONE

Dear Dr. Verfürden,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

ACADEMIC EDITOR

First, I must praise the authors for their extensive efforts in this work.

Second, some clarifications and improvements are still needed, as follows:

1- Are these figures produced by the same software mentioned in the methods, RevMan 5.3?

2- Why did you use the fixed effect model? I would advise using the random effect model instead owing to the present heterogeneity.

3- Publication bias assessment is not reliable in case of fewer than 10 included studies (according to egger et al). I advise you mention it only in the supplementary files (no need to be added to the full article).

4- I want to check the extracted data to repeat the analysis and compare with the reported results. Unfortunately, the raw data do not exist in the supplementary file. Please, provide them.

5-  When did you use MD and when did you use SMD? Why two formats of the Effect Estimate?

6- Did these studies report pre/post outcomes? Did you calculate the effect size based on (post - pre) or you consider the post values only for effect size calculation?

7- Cochrane Risk of Bias tool includes "blinding of study personnel", "blinding of outcome assessors" and "other bias" but I did not see these domains in the authors manuscript. You should include these domains in the ROB or clarify on-which basis they were omitted.

==============================

Please submit your revised manuscript by Sep 04 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Ahmed Negida, MD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In your Methods section, please provide additional information about the search performed (we note that the full search string is reported as a supplementary table, but would suggest that more information is included in the main text).

3. Please ensure that every statement is supported by an appropriate reference. For example, we note that the statement "No systematic review has previously evaluated the effects of LCPUFA on cognitive function in infant formula beyond early childhood (age �2.5 years) when cognitive tests are more accurate" is not followed by any citation corroborating the information on cognitive tests.

4. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

5.Thank you for stating the following in the Financial Disclosure section:

[This study was supported by funds from the Economic and Social Research Council, and the Great Ormond Street Hospital Charity. RG was supported by Health Data Research UK, an initiative funded by the UK Research and Innovation, Department of Health and Social Care (England) and the devolved administrations, and leading medical research charities. Research at UCL Great Ormond Street Institute of Child Health is supported by the NIHR Great Ormond Street Hospital Biomedical Research Centre. ].

We note that one or more of the authors have an affiliation to the commercial funders of this research study : Health Data Research UK

1. Please provide an amended Funding Statement declaring this commercial affiliation, as well as a statement regarding the Role of Funders in your study. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. You can update author roles in the Author Contributions section of the online submission form.

Please also include the following statement within your amended Funding Statement.

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

If your commercial affiliation did play a role in your study, please state and explain this role within your updated Funding Statement.

2. Please also provide an updated Competing Interests Statement declaring this commercial affiliation along with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc.

Within your Competing Interests Statement, please confirm that this commercial affiliation does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If this adherence statement is not accurate and  there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include both an updated Funding Statement and Competing Interests Statement in your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: N/A

Reviewer #4: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

Reviewer #4: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors performed a systematic review and meta-analysis of RCTs to assess the effect of LCUFA on long-term cognitive function in childhood. The authors concluded that “there was no evidence of benefit of LCPUFA supplementation in term or preterm-born infants and weak evidence that LCPUFA reduced IQ score in term-born children.”

Although the overall approach to the review is proper, I have few comments:

1. Abstract: Well-written structured abstract.

- I would recommend adding the software used for analysis.

- Please place the keywords after the abstract.

2. Introduction: The research question was clearly outlined, and the research question justified given what is already known about the topic.

- The text starts with references [3-5]. References should be numbered in consecutive order in the main text starting from “Introduction.”

3. Methods:

- The methodology described in the study seems to have followed current standardized procedures and guidelines for systematic review and meta-analysis. The methods are detailed enough to allow replication of the analysis. Outcomes are well-defined; the same is true for quality assessment and data analysis.

- Page 5: please define the PRISMA guidelines “Preferred Reporting Items for Systematic Reviews and Meta-Analyses.”

- Page 6: The paragraph starting with “The secondary outcomes…”, delete “standardised mean difference” because it is previously defined.

4. Results: The results are adequately presented.

- “We included eight unique trial cohorts,” please add references.

- a comma should be placed between the effect size and the 95% Cl.

5. Discussion: The results are discussed from multiple angles and placed into context without being overinterpreted, and conclusions answer the aims of the study.

6. Figure 1:

- The “Cochrane Central Register” search results are missing.

- 49 full-text articles assessed, the authors excluded 40 articles (27+13); thus, 9 articles should be retrieved, not 8. Please recheck!!

Reviewer #2: Abstract:

1- Please follow the guidelines of the journal (https://journals.plos.org/plosone/s/submission-guidelines)

Introduction:

2- References 1 and 2 are missing.

3- Please add reference for this information "Human breast milk contains DHA, AA, and their fatty acid precursors"

4- Define this abbreviation "IQ" at the first mention.

Methods:

5- Section of Methods need to be rearranged; you can develop a new sub-section for inclusion and exclusion criteria.

6- "Patient involvement" No need for this section or you can move it before the references.

7- Please move this section to be before the references "Role of the funding source"

Results

8- "We obtained previously unpublished outcome data for two RCTs: Firstly, a two-centre trial of term babies in England18 19 provided unpublished follow-up data on IQ assessments using the Wechsler Preschool and Primary Scales of Intelligence-Revised (WPPSI-R) at age 4.5 years and the Wechsler Abbreviated Scales of Intelligence (WASI) at age 16 years. Secondly, a two-centre trial in England of babies born preterm provided unpublished data from their IQ assessments using the WASI at age 16 years" Is their any duplication here? Please rephrase to avoid any confusion.

9- Please mention the Study ID (last name and year) in the table 1.

10- Remove any question mark in the table 1 and replace it with NA or NR.

11- In Figure 1: Please add additional arms for cochrane library and unpublished data

12- Did you used R program or Review Manager as these figures not belong to Review Manager?

13- Try to perform a sensitivity analysis to overcome this heterogeneity

Discussion:

14- The discussion is very week and need to be supported with some clinical aspects.

Reviewer #3: This is a systematic review and metaanalysis in which the authors investigated the long-term cognitive function effects of long-chain polyunsaturated fatty acids in infant formula in childhood at the age of 2.5 years or older. Although the study is well-written and structured, the overall quality of evidence for the included RCTs is low due to heterogeneity and quality of included studies. Moreover, many studies have investigated that topic extensively with similar results. I do not think that this paper will add more information to the medical literature related to that topic.

Reviewer #4: The authors conducted a meta-analysis to evaluate long-term effects of long-chain polyunsaturated fatty acids on IQ beyond 2.5 years of age. The findings are interesting and consistent with the previous literature. Some comments warrant attention:

* Abstract

- Why fixed effect model despite obvious heterogeneity?

- Add the number of randomized children in the results.

- weak evidence of lower IQ in the supplemented group: But the 95% CI is not consistent with this interpretation. It crosses the null value of zero for a quantitative MD.

- A single meta-analysis was registered on PROPSERO with two numbers!!

* What this study adds/What is already known: These sections are not suitable for PLOS One. It seems the manuscript was submitted using the author instructions of another journal (Similar is the abstract).

* Introduction

- There is an obvious problem with referencing: Numbers are not organized; many references are old back to 1992.

- Because human breast milk contains LCPUFA, could the authors account for breast-fed children in their analysis?

- At the start of the third paragraph, the authors should add "To our knowledge, ".

- The rationale for the meta-analysis is adequately stated.

* Methods

- "proceedings from major scientific meetings of child nutrition" can you specify?

- I see that the authors sub-grouped according to preterm/term delivery. But are there any data on children with low birth weight?

- "using validated measures" which included?

- The authors did not provide information on their data extraction process!

- "The primary outcome was the pooled difference, presented as mean difference (MD) and standardized mean difference (SMD), between intervention and control arms, based on the measure of cognitive function reported most frequently among the included studies" this part is not clear!

- So you selected fixed effect model based on previous meta-analyses?

- The secondary tests were cognitive test scores?? All IQ scoring tests, because there are various types of cognitive tests?

- The authors used random effects model only for secondary outcome bacause the studies used various tests. But also, the heterogeneity should have been expected for the primary outcome based on other different variables between studies.

- Add version and source of RevMan software! These forest plots do not appear to be extracted from RevMan!

- Also, specify that ou used the updated Cochrane ROB. II tool.

- The authors evaluated publication bias using funnel plots. They should mention that this method lacks reliability for less than 10 included studies.

* Results

- The authors should comment briefly on ROB assessment results of the included trials.

- What was the 2.5 year follow-up age cut-off based on?

- The authors analyzed based on total IQ scores. But these tests often have sections for different cognitive functions. Could the authors extract any data in this regard?

- Considering the I2 values in the forest plot, re-analysis under random effects model is needed!

- "The upper confidence interval included no difference (i.e. zero), but the average effect favoured a reduction in IQ in babies randomised to LCPUFA". So, I believe it should be interpreted as no difference as the CI is the imp value to determine significance here.

- For outcomes with substantial heterogeneity, what did the authors do to solve this heterogeneity?

* Discussion

- Well-written and comprehensive.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Reviewer #4: Yes: Abdelrahman Ibrahim Abushouk

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

15 Sep 2020

See attached point-by-point response document for the reviewers

Submitted filename: 2020-09-07 Response to reviewers.docx

Click here for additional data file.

21 Oct 2020

Effect of long-chain polyunsaturated fatty acids in infant formula on long-term ‎cognitive function in childhood: a systematic review and meta-analysis of randomised controlled trials

PONE-D-20-14949R1

Dear Dr. Verfürden,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Ahmed Negida, MD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

Reviewer #4: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #4: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #4: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #4: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #4: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have addressed all my comments/suggestions. I found their responses quite satisfactory and the revised version has been much improved.

Reviewer #2: I recommend accepting this manuscript, as the authors addressed all comments, and enhanced the manuscript significantly.

Reviewer #4: The authors have adequately addressed my concerns and I endorse the manuscript in the current for for publication.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Eshak I. Bahbah

Reviewer #4: Yes: Abdelrahman Ibrahim Abushouk

26 Oct 2020

PONE-D-20-14949R1

Effect of long-chain polyunsaturated fatty acids in infant formula on long-term ‎cognitive function in childhood: a systematic review and meta-analysis of randomised controlled trials

Dear Dr. Verfuerden:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Ahmed Negida

Academic Editor

PLOS ONE