A Dose–Response Meta-Analysis Between Maternal Fish Oil Supplement And Risk of Asthma/Wheeze In Offspring

Background: Prenatal exposure to omega-3 polyunsaturated fatty acids (n-3 PUFA) present in oily sh may prevent asthma or wheeze in childhood. Objective: By limiting this systematic review to sh oil intervention that commenced in the gestational period, we aim to nd more clear evidences about the relationship between supplement with sh oil during pregnancy and the risk of asthma/wheeze in offspring, and to improve the life satisfaction of children who suffered asthma. Methods: A comprehensive literature search was conducted in the following database: PubMed, Medline, Web of Science, the Cochrane library, and Embase up to February 2021. Two reviewers independently selected studies, extracted data of the characteristics, and assessed risk of bias. Eight randomized controlled trials totaling 3,037 mother-infant pairs were analyzed in the end. “Allergic asthma” and “asthma and/or wheeze” were assessed in our meta-analysis. Subgroup analysis and sensitivity analysis were conducted. Dose– response data was examined using the robust-error meta-regression method. Results: This meta-analysis showed that n-3 PUFA during pregnancy did not signicantly reduce the risk of asthma/wheeze (RR 0.93; 95% CI 0.82 to1.04, p=0.21) and allergic asthma (RR 0.66, 95% CI 0.24 to 1.86, p=0.44). Subgroup analyses revealed that the risk of childhood asthma/wheeze was signicantly decreased: (1) in Europe (RR 0.69; 95% CI 0.53 to 0.89), (2) when the dose was ≥ 1200 mg/d (RR 0.69; 95% CI 0.55 to 0.88), (3) when supplementation started after gestational age 22 (RR 0.65; 95%CI 0.50 to 0.85), (4) when supplementation was from pregnancy to lactation (RR 0.69; 95% CI 0.51 to 0.95). Furthermore, the linear dose–response analysis showed that when maternal supplementation of n-3 PUFA increased by 100mg/d, the risk of asthma/wheeze was reduced by 2%. Conclusions: Although perinatal replenishment of n-3 PUFA did not prevent allergic disease in offspring, under some conditions, it could reduce the incidence of asthma/wheeze and allergic asthma in children, and the higher the dose, the better the protective effect it has. Additional research is needed to conrm the hypothesis of a link between n-3 PUFA intake and prevention of childhood asthma/wheeze. and large sample size RCTs should be performed in the future to examine the effects of reasonable dose of prenatal n-3 PUFA intake and asthma/wheeze in offspring, especially in different races, location.


Background
As the most common allergic disease in childhood, the prevalence of asthma has increased rapidly in the past 20-30 years [1][2][3][4].
Many asthma children's life satisfaction had been affected, not only physical activity was limited, but also emotional and mental health had been impaired [5][6][7][8][9]. These effects create huge burdens on families, society, and medical care systems.
The modern diet life style, especially the balance of n-3 polyunsaturated fatty acids (n-3 PUFA), is widely accepted as a reason for an increased risk of allergic disease, although genetic, epigenetic, and environmental factors likely also contribute [10][11]. In many countries, high consumption of vegetable oils and meat results in an increased intake of n-6 PUFA and arachidonic acid (AA, 20:4, n-6), respectively. Diets high in n-6 PUFA lead to a high concentrations of AA in tissues. Arachidonic acid gives rise to prostaglandin and leukotrienes, both of which are highly active mediators of in ammation and allergic reactions [12]. Conversely, n-3 PUFA has multiple anti-in ammatory actions, such as reduce leucocyte chemotaxis, adhesion molecule expression and leucocyte-endothelium interaction, decrease production of in ammatory cytokines and the reactivity of T-cell, and increase production of eicosanoids with lower biological potency and in ammation resolving resolvins from eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) [13].
To obtain better clinical evidence, many systematic reviews have been performed. There are two systematic and meta-analyses [26,27] of observational studies assessing maternal sh oil intake in relation to risk of some allergic diseases, while the strength evidence of which was not better than results from randomized controlled trials. And there are three systematic and meta-analyses of RCTs [28][29][30]. Two of them evaluated maternal omega-3 PUFA intake in relation to risk of allergic disease in offspring [28,29], but the analysis on asthma/wheeze was not adequate. Another systematic review conducted by Lin et al. [30] found a protective effect of prenatal sh oil replenishment on wheeze/asthma in children, but the timeliness included in this meta-analysis is a little weak.
As described in this report, we undertook an updated systematic review to further measure a possible relationship between supplementation with omega-3 PUFA during pregnancy and the risk of asthma/wheeze in offspring, so as to improve the quality of life for children with asthma/wheeze.

Methods
This systematic review of RCTs was performed in conformity with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [31] that reported maternal sh oil intake during pregnancy and asthma/wheeze in children.

Study Selection
The meta-analysis included studies that met the criteria below: (1) Design: randomized controlled trial; (2) Participants: pregnant women and their children; (3) Fish oil group: supplement of capsules rich in n-3 PUFA or salmon commenced in the gestational stage; (4) Control group: supplement with placebo (e.g., olive oil, soya bean oil, vegetable oil); (5) Outcomes: prevalence of asthma/wheeze, and allergic asthma.
Two authors independently assessed articles for inclusion. Any discrepancies were resolved by discussion and, if necessary, by thirdparty arbitration.

Outcome measures
For this current review, the primary outcome was the prevalence of asthma/wheeze, de ned as either a clinical diagnosis, a parental report of asthma symptoms, at least three instances of wheezing in the previous two years, or parental report of physician diagnosis of asthma.
The secondary outcome was prevalence of allergic asthma, de ned as asthma with the presence of IgE antibodies or a positive skinprick test.

Data Extraction and Bias Assessment
Data were extracted using a standardized table. Extracted data included rst author with year of publication, study location, study design, participants, intervention, placebo, length of follow-up, and incidence of asthma/wheeze and allergic asthma.
The quality of studies was assessed using Cochrane Collaboration Risk of Bias Instrument [32]. A couple of authors investigators extracted the data from the selected studies and assessed quality of the articles, independently. If there were any differences or discrepancies, they would be settled through a third party discussion.

Data Synthesis and Analysis
Relative risk (RR) with 95% con dence interval (CI) was employed to assess the effects of n-3 PUFA supplementation during pregnancy on the incidence of allergic disease in offspring. Hazard ratio (HR) and incidence rate ratio (IRR) were directly regarded as RR. Odds ratio about the "allergic asthma" reported by Hansen et al. was considered as RR due to the lack of details [21]. When RRs were not reported in the article, we calculate the crude RRs according to the events/total of included studies. I 2 statistics were employed for quantifying the potential variability among included studies. When the heterogeneity was not obvious (I 2 ≤ 50%), the xed-effect model was implemented to summarize the results. When there was substantial heterogeneity (I 2 > 50%), the meta-analysis was undertaken by a random effect model [33]. Egger's test were conducted in this meta analysis to evaluate potential publication bias. Dose-response data using the robust-error meta-regression method was implemented by Stata 15.1/SE. A funnel plot was not executed because the number of studies was less than ten.
The above analyses and forest plots contained in this review were performed by the Stata 15.1/SE. The "risk of bias graph" was created by Review Manager 5.3.

Literature search
From the database search, we initially identi ed 3,310 publications from online-search in PubMed, Medline, Web of Science, Embase, and the Cochrane library. After deleting duplicate articles, 1,302 publications remained. Of that remainder, we removed 985 publications after reading the titles and 259 publications were removed after two independent reviewers read abstracts; this process resulted in 58 publications that required reading in full. Forty-eight full-text publications were rejected for various reasons. Finally, we selected 10 studies from 8 unique randomized controlled trials with the longest follow-up time for each outcome. Among them, 32 articles did not report the outcomes we needed, 8 articles were observational studies rather than RCTs, and 8 articles were not supplemented with n-3 long-chain PUFAs. Flow diagram named Fig. 1 elaborated the details of the study selection and progress of RCTs.

Participants
This review was composed of data for 3,553 women who had been pregnant, and 3,037 mother-infant pairs completed the randomized controlled trials. There were 1,603 women in the experimental groups and 1432 women in the control groups. All participants were recruited from clinics and hospitals. The selected studies were conducted worldwide, three studies in Australia [22,34,35], two randomized controlled trials from North America [25,35], and ve studies performed in Europe [20-21, 23-24, 37].
Participants in four randomized controlled trials [22][23][24]34] had an atopic disease or a family history of allergic disease. Two trials [20,21] included pregnant women who did not have medical problems. One trial included atopic and non-atopic mothers [36], and one trial recruited women with depression [25]. The length of follow-up was from 6 months to 24 years. Table 1 presents the characteristics and results of these randomized controlled trial studies. Table 1. RCTs of maternal n-3 PUFA supplementation during pregnancy and allergic disease in the offspring.

Intervention
In six of the eight unique randomized controlled trials, the participants were divided into an experimental group and a control group [20,[22][23][24][34][35][36], whose experimental group received salmon or sh oil, and control group received olive oil, vegetable oil, corn and/or soy oils, or nothing. In two trials [21,25], the participants were divided into three groups. In one trial [25], an eicosapentaenoic acid-rich sh oil group, a docosahexaenoic acid-rich sh oil group, and a control group (soy oil) were constituted. The other trial included [21] one experimental group ( sh oil) and two control groups (olive oil and no oil). The amount of daily n-3 PUFA supplementation was from 400 mg to 3700 mg.
The duration of intervention was from 10 to 29 weeks, after statistical calculation. 3.5. Quality of randomized controlled trials A bias risk assessment was performed employing the modi ed models of the Cochrane Collaboration Risk of Bias tool for intervention trials for the following six aspects, such as selection bias, performance bias, measurement bias, attrition bias, reporting bias, and other bias [32]. Individual item was assessed as low-risk, high-risk or unclear (not given).
The generation of random sequences was assessed as low-risk in all studies. Nine studies [20-23, 25, 34-37] showed ample allocation concealment, and one study [24] did not describe the allocation concealment; thus, it was not possible for the experimental subjects and researchers to predict the results of the study. The pregnant women and medical staff were blinded in seven studies and unblinded in three studies. In the unblinded studies, the pregnant women in the control group did not receive any supplement [ while not in the others [20,22,[24][25]35]. Five studies [20-21, 34, 36-37] showed a low risk of reporting bias, and four studies had high risk of reporting bias [22-23, 25, 35]. In nine studies, there was not enough information to evaluate whether there were other risks of bias or whether current problems introduced bias, and only one study was rated as a low risk of other bias because of no obvious biases in the report [23]. Figure 2 shows the risk assessment of bias for all the studies 3.6. Meta-analysis results of the trials Asthma/wheeze In the clinical practice, it is di cult to explicitly diagnose "asthma" in children because of the strict diagnostic criteria of asthma compared with wheeze [38,39]. Therefore, we chose the prevalence of asthma/wheeze as the main outcome to evaluate the effect of sh oil intake during pregnancy. Four reports were from two randomized controlled trials; thus, we assessed eight studies with longer follow-up for the incidence of asthma/wheeze. Six trials did not reveal any signi cant differences for asthma/wheeze between the sh oil group and the placebo groups [22][23][24][25][34][35]. Two trials showed signi cant protective effects of the intervention during pregnancy [20,21]. The pooled data of the eight trials [20][21][22][23][24][25][34][35] showed that consumption of n-3 PUFA during pregnancy did not have a signi cant protective effect against asthma/wheeze compared with placebo (RR 0.93; 95% CI 0.82 to1.04; p = 0.21) (Fig. 3-A).

Allergic asthma
In three randomized controlled trials, investigators reported the effect of n-3 PUFA intake on the incidence of allergic asthma [21][22]34], and only one trial showed a signi cant association between maternal sh oil intake and childhood allergic asthma [21]. The pooled data of three trials [21][22]34] indicated that the prevalence of allergic asthma in children was not reduced in experimental group compared with the placebo controlled group (RR 0.66, 95% CI 0.24 to 1.86, p = 0.44; Fig. 3-B).

Subgroup analysis
We analyzed the outcomes by strati ed study location, family history of allergic disease, gestational stage when supplementation commenced, duration of supplementation, and age of offspring.
The risk of our main outcome endpoint in offspring, asthma/wheeze, was notably decreased under the following conditions: (1) in Europe, (2) with ≥ 1200 mg/d, (3) when gestational age supplementation started after 22 weeks, (4) and when supplementation continued into lactation. Table 2 shows the detailed results of asthma/wheeze subgroup analysis. We found that prenatal n-3 PUFA supplementation could decrease the incidence of allergic asthma in preschool children (aged 6 years or less). Table 3 presents the detailed information about the age subgroup of allergic asthma. Table 2 Subgroup analyses of n-3 PUFAs supplementation during pregnancy on the incidence of asthma / wheeze.  The results showed that there was a linear dose-response relationship between the daily sh oil supplement dose during pregnancy and the incidence of asthma/wheeze, and the higher the dose, the lower the incidence, and when perinatal n-3 PUFA supplementation increased by 100mg/d, the risk of asthma/wheeze was reduced by 2%. Details of linear regression model was in the following Fig. 4, and other detailed data and code of Stata on doseresponse analysis was shown in Appendix 2.

Publication bias
No statistically signi cant publication bias was suggested by the outcome of Egger's test of asthma/wheeze (p = 0.54) and allergic asthma (p = 0.34) for each outcome in this meta-analysis.

Discussion
The current systematic review did not show an explicit relationship between prenatal intake of n-3 PUFA and the prevalence of asthma/wheeze in all offspring, which was similar to a recent study by Vahdaninia et al. [29] and as summarized in a review by Best et al. [28]. Differently, Lin et al. reported a protective effectiveness of prenatal sh oil replenishment on wheeze/asthma in children in his meta-analysis [30], while two RCTs [23,36] in which were not included, especially one large-sample RCT performed in Mexico [36].
The result of subgroup analysis was interesting. Studies in Europe showed a protective effect on wheeze/asthma in offspring. For children without family histories of allergic diseases (i.e., medically diagnosed allergic disease, e.g., eczema, asthma, or hay fever), maternal supplementation with sh oil in gestational period did not show any advantage for prevention of asthma/wheeze after removal of a study conducted in Mexico, a large sample trial included in our review (at the same time, the heterogeneity decreased from 64-0%; Appendix 3). We attribute those changes to differences in ethnicity and environment. And the association between the prevalence of allergic diseases and race/ethnicity has also been con rmed in other studies [40][41][42][43]. Further research is needed to explore whether sh oil replenishment during pregnancy bene ts children in an ethnicity-dependent manner.
Compared with the effective dose of 2000 mg/d reported in previous reviews [30], we found that only 1200 mg DHA and/or EPA per day signi cantly prevented asthma/wheeze. Further studies should more thoroughly examine by detailed dose-response meta-analysis the effective dose of maternal sh oil intake.
Supplementation of sh oil that began after 22 weeks gestational age and continued into early lactation signi cantly reduced the prevalence of asthma/wheeze in offspring. We hypothesized that n-3 PUFA might have different activities in embryo development at different stages of pregnancy because pro-in ammatory immune cell genes are expressed in late pregnancy, and the period of late pregnancy has a key regulatory function in in ammatory and immune system development [44,45]. The immune system of a newborn is highly malleable. If the immune system does not receive appropriate signals, the neonate will incur susceptibility to allergic diseases [46][47][48]. Therefore, we speculate that there is a "window of opportunity" in early life during which the immune system may be acted upon by sh oil and thereby limit susceptibility to allergic diseases.
Phenotypes of asthma in children are commonly associated with allergy, and the incidence of allergic asthma will decline to some extent with advancing age [49,50,51]. We found supplementation of sh oil reduced the prevalence of allergic asthma in preschool children (< 5y) in subgroup analysis, but we did not nd any relevant studies to explain this phenomenon. More research is needed to measure the relationship between maternal sh oil supplement and the onset of allergic asthma in childhood.
Our meta-analysis had several advantages. All included studies were recent randomized controlled trials with large sample sizes. We performed subgroup analysis and sensitivity analysis to assess potential confounding factors and evaluate stability of the outcomes.
The main strength of this systematic review was that we used the most precise de nition and the broadest de nition of asthma --"allergic asthma" and "asthma/wheeze" as our outcome variables. However, because it was based on randomized controlled trials, our systematic review had some limitations. The primary studies had different protocols that may have affected our ndings to some extent. For example, the baseline characteristics of the pregnant women were different, and the dose of intervention and the diagnosis of outcomes were also different.
Because of a lack of new and large-sized randomized controlled trials, possible confounding factors and potential bias, the hypothesis that linking maternal n-3 PUFA intake to protection against childhood asthma/wheeze or allergic asthma cannot be accepted or rejected absolutely. Ethnicity, dose, susceptibility of n-3 PUFA, and time of supplementation should be assessed in the future. Largesample, multi-center, and randomized controlled trials need to be conducted to better comprehend the e cacy of supplementation with n-3 PUFA in pregnancy for protecting against asthma/wheeze or other relevant allergic diseases.

Conclusion
The available evidence showed prenatal replenishment of n-3 PUFA may reduce the incidence of asthma/wheeze or allergic asthma in offspring under certain conditions, and the dose-response analysis showed that the higher the dose, the stronger the protective effect it has. More high-quality and large sample size RCTs should be performed in the future to examine the effects of reasonable dose of prenatal n-3 PUFA intake and asthma/wheeze in offspring, especially in different races, location.

Declarations
Ethics approval and consent to participate: Not applicable.
Consent for publication: Not applicable.
Availability of data and materials: The datasets analyzed in this study are available from the corresponding author on reasonable request.
Competing interests: The authors declare that they have no competing interests. Flow diagram depicting the study selection and progress of RCTs identi ed in the systematic review and meta-analysis. RCT, randomized controlled trial.

Figure 2
Assessment of risk of bias for included RCTs.

Figure 3
Effect of n-3 LC-PUFA supplementation during pregnancy compared with placebo on the incidence of asthma and/or wheeze (A) and allergic asthma (B) of children. The pooled estimate was obtained using a xed-effects model depending on the heterogeneity test. Squares represent RRs and error bars represent 95% CIs. The diamond represents the overall effect estimate. The size of the shade square is proportional to the percent weight of each study.