OFFSPRING: A SPRING Follow-Up Study Assessing the Efficacy of Maternal Probiotics and Allergic Disease in the Child

Abstract

Introduction

There are a variety of factors that contribute to the development of allergic diseases in children, including environmental exposures during the maternal prenatal period. It has been proposed that probiotic supplementation during pregnancy could be used as a possible preventative measure to target childhood allergic disease.

Methods

Participants from a previously conducted prospective double-blind randomised control trial of probiotics versus placebo study (Study of PRrobiotics IN Gestation) were sent electronic questionnaires to complete about their child, who are now between 3 and 7 years of age. Demographic data and rates of allergic diseases were compared between the two groups.

Results

One hundred and seven women responded to the questionnaires. Between the two groups, there was no difference in the frequency of allergic diseases, with similar rates of eczema, asthma, and hospital presentations seen.

Conclusion

In this follow-up study, infants of mothers who were exposed to probiotics during their pregnancy do not appear to have any paediatric health advantages in terms of allergic diseases.

Introduction

Allergic diseases are some of the most common non-communicable chronic diseases in children and contribute to a significant health and economic burden in both developed and emerging countries, with at least 30% of the population being affected [1]. Significantly, the prevalence of allergic diseases continues to increase globally [2, 3]. Allergic diseases include asthma, allergic rhinitis/hay fever, eczema, as well as food, insect, and drug allergies [4]. Although data on the true global economic costs of allergic diseases are not well documented, it is estimated that asthma alone costs the US economy more than USD 81 billion annually [5].

It is widely accepted that there are a variety of factors that contribute to the development of allergic diseases, including genetics, environmental factors, dietary habits, antibiotic use, and housing conditions [6–8], with some evidence suggesting that environmental exposures during the prenatal period may also impact upon the unborn child [7, 9]. Evidence suggests that the composition of the microbiome in the foetus may influence early life immune programming and subsequent risk of immune-mediated diseases (including allergic diseases) throughout a person’s life [7].

As a strategy to reduce increasing rates of allergic diseases, it has been proposed that probiotic supplementation during pregnancy could be used as a possible preventative measure targeting childhood allergic diseases [10], through modulation of the gut microbiota, which can impact on immune programming and disease risk [11]. The World Allergy Organization (WAO) currently recommends the use of probiotics in pregnant women at high risk of having an allergic infant as “there is likely net benefit” from probiotic use [10]. However, the guidelines acknowledge that the evidence supporting a reduction of allergic diseases in infants whose mothers took probiotics during pregnancy is of very low quality, with a systematic review of randomised trials finding many inconsistencies and imprecision of results in current research [12]. Given the paucity of data to support the effects of probiotics administered during pregnancy, the WAO also acknowledges that long-term follow-up of well-designed randomised controlled trials of probiotic supplementation in pregnant women is required to address this knowledge gap [10].

Aims

The aim of the study was to use offspring follow-up data from a randomised controlled trial evaluating probiotic supplementation in pregnancy for the prevention of gestational diabetes as the primary outcome, to provide information about the efficacy of maternal probiotic supplementation during pregnancy on the subsequent health, growth, and allergic disease development in the child.

Study of PRrobiotics IN Gestation

Study of PRrobiotics IN Gestation (SPRING) was a prospective double-blind randomised controlled trial of probiotics (Lactobacillus rhamnosus GG + Bifidobacterium lactis BB12) versus placebo, which primarily assessed the impact of probiotics on maternal gestational diabetes mellitus in overweight and obese women. Participants of the SPRING study were recruited at less than 16 weeks gestation and were required to be pregnant with a singleton pregnancy and have a BMI of >25.0 kg/m². Exclusion criteria included women with known pre-existing diabetes or gestational diabetes diagnosis prior to recruitment, maternal medical condition associated with altered glucose metabolism, or known foetal abnormality on 12th week ultrasound examination. Detailed methodology and information about this study have been previously published [13], with results finding that probiotic use from the second trimester of pregnancy did not prevent gestational diabetes mellitus [14]. Children born to mothers who participated in the SPRING study were between 3 and 7 years of age when this follow-up study commenced, providing an opportunity to assess whether maternal probiotic supplementation in pregnancy is associated with lower rates of childhood allergic diseases. Secondary outcomes included assessing the frequency of other medical conditions in the child, as well as the child’s growth.

Methods

Women who participated in the SPRING study were contacted via email to offer participation in the OFFSPRING follow-up study. Women who were uncontactable by email were then phoned to offer an opportunity to update their contact details. The inclusion criteria were children born to mothers who participated in the SPRING study and who had previously consented to further contact from the research team. Women whose child was born with a major anomaly, or who had died, were not contacted for follow-up (n = 3).

Data Collection

Data were collected through five detailed questionnaires which were emailed to women. The questionnaires were sourced from the Barwon Infant Study [15] and assessed a wide range of health outcomes, including the child’s general health, growth and development, illness, hospital admission, and allergic diseases (including asthma, eczema, allergic rhinitis, food and nut allergies, and anaphylaxis), as well as maternal physical and mental health.

Health outcomes were self-reported by the mother of the child born from the SPRING study, including their assessment of their child’s general health (excellent, very good, good, fair, poor), number of hospital presentations and admissions, and number of cold and flu lasting 3 or more days in the past 12 months. Criteria applied to assess food allergies, eczema, asthma, hay fever, attention-deficit/hyperactivity disorder, autism spectrum disorder specified a diagnosis from a doctor (e.g., “Were you told by a doctor that your SPRING child had a food allergy?”).

Analysis

Baseline characteristics of mother and child were summarised, comparing those who did and did not respond (Table 1). Of those who responded, characteristics of the probiotic and placebo groups were compared (Table 2). Characteristics were summarised as n (%) for categorical variables, and p values were calculated. Data were analysed using Stata Statistical Software: Release 18 [16]. Analyses were conducted by LC.

Table 1.

Comparison of baseline demographic characteristics of SPRING women who did and did not complete follow-up questionnaires

	Followed up (N = 107), n (%)	Not followed up (N = 272), n (%)	p value
Age at baseline, mean (SD), years	32.1 (4.2)	31.3 (5.0)	0.09
Assigned probiotic arm	52 (48.6%)	142 (52.2%)	0.52
Caucasian	97 (90.6%)	230 (84.5%)	0.12
Hospital
Mater Hospital	48 (44.8%)	122 (44.9%)	0.25
Royal Brisbane & Women’s Hospital	54 (50.5%)	124 (45.6%)
Redcliffe Hospital	5 (4.7%)	26 (9.5%)
Diagnosed with GDM	15 (14.0%)	45 (16.5%)	0.56
Diagnosed with HDP	11 (10.3%)	43 (15.8%)	0.43
Tertiary educated	62 (57.9%)	119 (43.7%)	0.013
Smoker	0 (0%)	12 (4.4%)	0.03
Nulliparous	46 (43.0%)	104 (38.2%)	0.39
BMI at baseline, mean (SD), kg/m2	33.1 (6.6)	32.9 (5.6)	0.64
Systolic BP, mean (SD), mm Hg	98.6 (10.4)	101.4 (5.78)	0.40
Diastolic BP, mean (SD), mm Hg	58.06 (10.0)	59.01 (5.55)	0.46
Previous GDM	4 (6.5%), n = 61 multiparous	18 (10.7%), n = 168 multiparous	0.34
Family history of diabetes	21 (19.6%)	77 (28.3%)	0.08
Chronic hypertension	4 (3.7%)	6 (2.2%)	0.40
Alcohol use during pregnancy	9 (8.4%)	11 (4.0%)	0.09
Gestational age at delivery, mean (SD)	39.2 (2.0)	39.2 (1.92)	0.85

GDM, gestational diabetes mellitus.

Table 2.

Maternal and infant characteristics following probiotics or placebo

Maternal characteristics	Probiotic (n = 62), n (%)	Placebo (n = 62), n (%)	p value
Age (SD)	37.0 (4.8)	37.9 (4.2)	0.87
Nulliparous at baseline	28 (45.1%)	28 (45%)	1.0
Caucasian	57 (91.9%)	54 (87.1%)	0.38
GDM during SPRING pregnancy	5 (8.1%)	7 (11.2%)	0.56
HDP during SPRING pregnancy	6 (9.6%)	5 (8.1%)	0.58
Tertiary educated	41 (66.1%)	38 (61.2%)	0.57
Current smoker	1 (1.6%)	0 (0.0%)	0.31
BMI (during pregnancy), mean (SD)	32.6 (7.8)	32.9 (6.0)	0.43
Current BMI, mean (SD)	32.3 (7.0)	33.1 (6.4)	0.26
Current maternal mental health	15 (24.1%)	14 (22.5%)	0.83
General health (very good, excellent)	23 (37.1%)	29 (46.7%)	0.5
Current allergic diseases (asthma, eczema, food allergies, hay fever, rhinitis)	32 (51.6%)	36 (58.1%)	0.47

Offspring characteristics	Probiotic (n = 52), n (%)	Placebo (n = 55), n (%)	p value
Age	5.07	5.04	0.59
Female	26 (50.0%)	28 (50.9%)	0.92
BMI, mean (SD)	20.53 (4.8)	20.31 (4.6)	0.58
Gestational age at delivery, mean (SD)	39.1 (2.1)	39.3 (1.7)	0.58
General health (very good, excellent)	44 (84.6%)	47 (85.4%)	0.90
Food allergy	7 (13.4%)	7 (12.7%)	0.91
Eczema	14 (26.9%)	13 (23.6%)	0.696
Wheeze and asthma	8 (15.3%)	9 (16.3%)	0.89
ADHD	2 (3.8%)	0 (0.0%)	0.14
ASD	2 (3.8%)	3 (5.4%)	0.69
ED presentations	35 (67.3%)	39 (70.9%)	0.68
Hospital stays	11 (21.1%)	12 (21.8%)	0.93
>4 colds/flus per year	27 (51.9%)	20 (36.3%)	0.19
Antibiotics in the last 12 months	24 (46.1%)	21 (38.1%)	0.40

GDM, gestational diabetes mellitus; ADHD, attention-deficit/hyperactivity disorder; ASD, autism spectrum disorder.

Results

The SPRING study consisted of 411 women – 204 in the placebo group and 207 in the probiotics group. Of these 411 women, 32 had not consented to further follow-up studies and therefore were not contacted. Questionnaires were emailed to 379 women.

Maternal Characteristics

Out of 379 women, 107 responded to and emailed the questionnaires (28.2%). Women were more likely to respond to the questionnaires if they were tertiary educated (p = 0.013) or a non-smoker (p = 0.03), based on previously collected information from the SPRING study (Table 1). Of the women who responded to the questionnaires, there were no differences in maternal characteristics between probiotic and placebo groups (Table 2).

Offspring Characteristics

Characteristics of children in the probiotic and placebo groups were similar in terms of age, gender, and BMI. When comparing results from those who responded to the questionnaires in probiotic and placebo groups, there were no differences in any childhood outcome. In particular, there was no difference in the frequency of allergic diseases, with rates of eczema, asthma, and hospital presentations similar between both groups.

Discussion

This study demonstrates that infants whose mothers were exposed to probiotics containing L. rhamnosus GG and Bifidobacterium animalis subsp. lactis BB-12 from early second trimester and throughout their pregnancy do not appear to have any paediatric health advantage in terms of allergic diseases into childhood, when compared to those whose mothers did not receive a probiotic during their pregnancy.

Previous systematic review’s findings have indicated that the effect of maternal probiotics on childhood allergic diseases is inconsistent, with the need for larger scale studies to be conducted to investigate the potential effect more thoroughly [17]. Similarly, difficulties remain with interpretation of various study results due to the heterogeneity in the species and strain of probiotic used, dose, duration, and timing of use [18]. Mechanistic studies in mice have shown a role of maternal ability to produce short-chain fatty acid by fermentation of dietary fibres in the programming of regulatory T cells, which can reduce the predisposition for asthma [19]. However, the probiotics used in this study do not alter short-chain fatty acid production, which could further explain the lack of effect in the study participant’s offspring. Studies that demonstrated positive effects from probiotic use report administration in both prenatal and postnatal periods, suggesting that continued supplementation to the baby may contribute to reduced rates of allergic diseases [20]. Additionally, the pregnant cohort who participated in the SPRING study were not at higher risk for having children with allergic diseases. For those children at an increased risk of developing allergic diseases, that is, who have a first degree relative with allergic disease, probiotic supplementation during pregnancy may yield a more beneficial outcome, when compared to the general population [21].

As found with other follow-up studies, challenges were noted by research staff in terms of a lack of resources – namely, time and funding for this research to occur. Planning and preventing loss to follow-up are vital methodological considerations as part of the initial implemented study [22]. Participants being lost to follow-up can potentially compromise internal and external study validity, particularly when biases due to differences in those who respond to follow and those who do not are noted [23]. Within the cohort of those who participated in this follow-up study, characteristics are comparable to those who did not participate, which provides some confidence this bias did not occur.

This study has a number of limitations. Participants of the SPRING study birthed in 2013–2016, and therefore, there had been an extended length of time since women had been contacted by study staff. Consequently, many women had changed their contact details and even though email “bounce backs” were not commonly noted when questionnaires were sent, it is assumed that some participants’ emails were no longer in use. Of significance, this study highlights the importance of funding for follow-up studies from the outset. This would allow for follow-up research to be conducted, including paying research staff to update participant’s details and contact participants to encourage completion of questionnaires.

Due to the small portion of women who responded to questionnaires, this study is likely to be underpowered and we cannot exclude an effect. However, the findings are important to share with the scientific community and remain relevant, adding to developing evidence within this field. Similarly, there is the potential for selection bias given that over two-thirds of women from the SPRING study did not choose to participate in this follow-up study. Equally, our study demonstrates that in women who do not have a history of allergic diseases, there are not an overwhelming paediatric benefit identifiable and also no signal of harm.

Conclusions

Children of mothers who were randomised to have probiotics in pregnancy did not demonstrate any improvement in rates of allergic diseases, including asthma, eczema, food allergies, or hospital presentations. This study highlights the barriers of maintaining participant engagement for future follow-up studies. Larger randomised trials that are adequately funded from the outset are required to further investigate the potential positive or negative effects of probiotic use in pregnancy for childhood health outcomes.

Acknowledgments

We thank all families for their continued participation in the OFFSPRING study. We also wish to thank Professor Peter Vuillermin and Professor Ann Louise Ponsonby for permission to use their Barwon Infant Study Questionnaire for the purpose of this research.

Statement of Ethics

The study was approved by the Human Research Ethics Committees of the Royal Brisbane and Women’s Hospital (HREC/11QRBW/467). Participants of the SPRING study who consented to be contacted for follow-up research projects were contacted via email or phone, provided with written information about the study and the opportunity to discuss their questions with research staff. Questionnaires were emailed to the participants, who were informed in written information that completing the questionnaires indicated consent to participate in the study. This study was conducted ethically in accordance with the World Medical Association Declaration of Helsinki.

Conflict of Interest Statement

The authors have no conflict of interests to declare.

Funding Sources

This study received financial support from a RBWH Foundation Grant.

Author Contributions

L.C. designed the study and undertook the statistical analysis. K.F. and E.V.S. were responsible for data collection. L.C., E.V.S., K.F., M.D.N., H.B., H.D.M., and M.T. contributed to the interpretation of the data. E.V.S. wrote the manuscript. All authors critically revised the manuscript.

Funding Statement

This study received financial support from a RBWH Foundation Grant.

Data Availability Statement

All data generated or analysed during this study are included in this article. Further enquiries can be directed to the corresponding author.