Preventive effect of prenatal maternal oral probiotic supplementation on neonatal jaundice (POPS Study): A protocol for the randomised double-blind placebo-controlled clinical trial

Study design, setting and period

This is a randomised double-blind placebo-controlled parallel group superiority clinical trial in the Prince of Wales Hospital (PWH), Hong Kong from January 2024 to June 2025. This trial will be with pregnant women randomly assigned to either the probiotic group (Vivomixx with maltose) or the placebo group (maltose only). The trial protocol is reported as per the Standard Protocol Items: Recommendations for Interventional Trials guidelines.62

Participants

Pregnant women who will attend their antenatal care and plan to deliver in PWH will be invited to participate in this study.

Inclusion criteria

• Pregnant women from 18 to 45 years old.

• Gestational age between 28 and 35 weeks.

• Normal singleton pregnancy.

• Not on standing dose of antibiotic treatment during enrolment and initiation of the intervention.

• Plan to exclusively breastfeed.

Exclusion criteria

• Pregnant women with any fetal abnormality.

• Couple with glucose 6-phosphate dehydrogenase enzyme deficiency.

• Couple with known rhesus or haemolytic disease history.

• Any known risk of developing pathological jaundice.

• Plan to give birth at other hospitals than PWH.

• Women with any contraindications for breast feeding.

• Non-exclusively breastfeeding babies.

Recruitment and randomisation

Pregnant women will be recruited between 28 and 35 weeks in the outpatient clinics in a tertiary-level hospital (figure 1). After confirming the eligibility of pregnant women, the investigator will explain the study by using a printed information sheet that contains the title of the study, purpose, expected length of time for participation, description of all the procedures, confidentiality, voluntary participation, the right to refuse the intervention, withdraw from participation in the clinical trial at any time, contact information and all information will be kept securely. Sufficient time is allowed for their consideration, and all their questions will be answered. Pregnant women will voluntarily sign the written informed consent form (online supplemental file 1). Computer-generated randomisation and allocation will be made by a third party other than the investigators and clinicians to allocate the pregnant women to either probiotic or placebo groups in (1:1) ratio. Neither the study participants nor researchers will know the randomisation and allocation. The intervention will be initiated at 36 weeks of gestation for both study groups. To date, the recruitment has not started yet.

Figure 1

Flow diagram of participant recruitment and follow-up in the POPS trial using Consolidated Standards of Reporting Trials. DOL, days of life; N, total screened for eligibility; n, number of participants in each stage of enrolment, intervention and follow-up, POPS, Perinatal Oral Probiotics Supplementation.76

Intervention

The probiotic group pregnant women will take one sachet of probiotic (a maltose containing Vivomixx) product orally daily from 36 weeks of gestation up to the seventh day of postpartum while pregnant women in the control group will take one sachet placebo (only maltose) orally daily for the same duration. The best time to take the product is in the morning before breakfast. Each participant in both groups will be advised to dilute a sachet of product provided with a cap of cold water. Detailed information will be provided to each participant with an information sheet prepared in local language. Both active Vivomixx and placebo are similar in colour and taste with the same white sachet package. The probiotic product in this trial is a widely studied supplement among pregnant women for other research outcomes including vaginal microbiota and cytokine secretion, bacterial vaginosis, breast milk beneficial microbiota and cytokines, glycaemic control, neonatal gastrointestinal functional symptoms and gut microbiota as well as immune response.54–59 Maximum gut microbiome colonisation is achieved in 2–3 weeks of supplementation.63

Investigators have trial identified about 12 different microbiome species from the literature that have direct or indirect bilirubin reduction effects (table 1). There are commercially available probiotic products that contain many of these species. Vivomixx is one of the accessible well-studied products for its safety during pregnancy. It is a non-GMO, gluten-free, high-potency microbiotic food supplement, containing eight strains of live bacteria (450 billion bacteria per sachet) including S. thermophilus DSM24731/NCIMB 30438, Bifidobacterium breve DSM24732/NCIMB 30441, Bifidobacterium longum DSM24736/NCIMB 30435, Bifidobacterium infantis DSM24737/NCIMB 30436, Lactobacillus acidophilus DSM24735/NCIMB 30442, Lactobacillus plantarum DSM24730/NCIMB 30437, Lactobacillus paracasei DSM24733/NCIMB 30439, Lactobacillus delbrueckii ssp. bulgaricus DSM24734/NCIMB 30440.64 From lines of affidavits, we have cross-evaluated and found that seven of the strains affect the bilirubin metabolism. It can be stored at room temperature for up to week without having a major effect on potency. Otherwise, it is safe at 4°C–8℃.65 66 We prepared an icebag that can keep the product at a low temperature during transportation to home and each participant will store the product in a refrigerator at home. It will be stored, distributed and administered as per the product instruction by clinicians.

Outcome measures

Secondary outcomes

Secondary outcomes involve (1) Pregnancy outcomes will be collected by a questionnaire which contains different outcome measure variables including fetal well-being, preterm labour, pre-eclampsia, diabetes mellitus, infection/sepsis, induction of labour, mode of delivery, GA at birth, birth weight, Apgar score and …etc. (2) Milk and stool microbiome: Breast milk samples will be collected within the second and on the seventh day of giving birth. It will be kept at 4°C during transportation and stored at 80°C freezer. The maternal stool will be collected before initiation of the intervention (at 36 weeks of gestation) and after 3 weeks of intervention. Infant stool will be collected simultaneously with breast milk samples (figure 2). Both maternal and infant stool samples will be collected using a tube containing DNA stabiliser and then aliquoted and frozen at 80°C. Both samples will be processed based on the commercially available kit protocol. Metagenomic sequencing will be used to characterise the microbiome in each sample. For analytical measurement, the number of participants from whom provided probiotics were detected (relative risk) and the mean concentration of provided probiotic species in the breast milk and infant stool will be measured. Maternal stool sample microbiome profiles before and after intervention in both probiotic and placebo groups will be compared. (3) Safety for the mother and fetus/ infant: An independent safety monitoring committee and all research team members will monitor each participant for any adverse events, in particular, gastrointestinal symptoms and possible allergies. Mothers and infants will be followed regularly for possible side effects such as allergic reactions, stomach upset, diarrhoea, flatulence (passing gas) and bloating.37 Apart from discomfort causing symptoms such as bloating other severe side effects are uncommon. However, in any case, such a moderate to severe side effects, such as decreased/absent/fetal movement, fever, uncommon vaginal fluid leakage, severe headache, decreased abdominal girth or overdistended abdomen, the intervention will be stopped, the participant will be linked to the Accident and Emergency Department. If any adverse events occur in infants, infants will be linked to the paediatrics department.

Figure 2

Timeline of the POPS study from recruitment to end of follow-up. POPS, Perinatal Oral Probiotics Supplementation

Follow-up, compliance and dietary intake data

Pregnant women will be reminded to collect preintervention and postintervention stool samples, start the intervention at 36 weeks of gestation and be advised to report when labour starts or any discomfort. For better retention of participants, a reminder message will be sent to each participant using WhatsApp in the local language at four different time points. The pregnant women will be reminded to record daily on the diary form regarding the dose and time they take the supplement, supplemented with any use of prescribed antibiotics or adverse events such as any diarrhoea, hospitalisation, etc. In addition, they will also be reminded to keep their usual diet only and not to take any other probiotic products. The unused supplements report will be collected. Maternal dietary intake and diversity will be evaluated using standard individual dietary diversity measurement guidelines.75 76

Data and sample collection

For this randomised controlled clinical trial, data will be collected at least at three time points.

• Recruitment: At this stage, participants’ characteristics such as age, marital status, educational level and other sociodemographic characteristics, obstetrics and gynaecological characteristics, medical and surgical factors, reproductive history-related factors including family size, behavioural-related factors (smoking or alcohol use), allergies, use of supplements/vitamins/probiotics, prior lactation history, use of standing dose of antibiotics, plan for the place of birth, current pregnancy condition (gestational age, type of pregnancy (singleton), presentation, fetal heartbeat, placental location, fetal weight, any supplement/ vitamin/ probiotic in the previous 1 month. Preintervention maternal stool samples will be collected from women in both groups. The intervention will be initiated at 36 weeks of gestation and after 3 weeks of intervention, postintervention data will be collected.

• Within 2 days of life and before discharge:

  Having sufficient rest after delivery, mothers will be contacted until before discharge home for data collection at their most convenient time. The time for data collection will be marked. Data including gestational age at delivery, date and time of delivery, mode of delivery, any antibiotic use; infant birth weight, Apgar score, urination status of the infant and time of initiation of breast feeding will be documented. Breast milk and infant stool samples will be collected, and the volume of samples collected, and storage temperature for both samples. Moreover, the neonate’s TcB level will be measured and documented.

• Follow-up on the seventh day of infant life: Breast milk and infant stool samples will be collected. Both the volume and storage temperature of samples will also be recorded. Infant body weight and bilirubin level will be measured. Breastfeeding status will be recorded. The diary record for supplements will be returned and the final sachet count report will be recorded.

In general, the data collection will be exclusively by the research staff appointed for this research using a pretested questionnaire. After recruitment, every mother will be counselled at each contact for better compliance with the intervention.

Data handling and quality assurance

All the data collected through this study will be handled by the principal investigator and researchers and kept confidential. Data quality will be checked regularly by all investigators separately and a trial managing committee.

Sample size

The sample size in this protocol is calculated using the G*power V.3.1.9.4 statistical software.77 Due to the lack of previous studies on pregnant women targeting bilirubin level reduction in neonates, we used a previous study conducted to assess the effect of a probiotic on neonatal hyperbilirubinaemia by providing the lactobacillus RGG directly to neonates in Turkey to calculate our study sample size (PMID: 30968632).78 The effect size was calculated using mean, SD and sample size, which are the most common parameters to measure effect size. In the paper conducted in Turkey, the mean and SD of bilirubin level and the sample size were mentioned, and we took the third-day bilirubin measurement records. The effect size was computed as 0.663. Finally, the sample size is calculated using the following assumptions: Two tailed, power 80, effect size 0.663, α=0.05, 1:1 allocation ratio. This yields about 74 (37 in each arm), and then we considered a 20% drop-out using the formula N=(n/1−DO). The final sample size, therefore, computed 94 (47 in each group).48 We have also calculated the sample size using a reduction of the incidence of neonatal jaundice in controls (p=46.6%) and in the probiotics group (p=14.6%) and yield (n=62; 31 in each group), which is smaller than the first sample size. To calculate this sample size, we used additional parameters (α=0.05, power=0.8, enrolment ratio 1:1) and calculated using an online platform.79

Statistical analysis

This study will use the Consolidated Standards of Reporting Trials guidelines.80 An intention-to-treat approach will be used to assess the cause-and-effect relation to maintain the effect of randomisation. Baseline characteristics will be compared between allocated treatment groups using frequencies and descriptive statistics. Pearson’s χ² test and Fisher’s exact test will be run to assess the difference of different factors between the two intervention groups (online supplemental table). To assess the exclusive effect of the intervention on the occurrence of neonatal jaundice, a binary logistic regression model will be fitted. Additionally, to examine the effect of the intervention on the mean bilirubin level of neonates, mean beneficial bacteria level in breast milk, and in maternal and infant stool, linear regression will be employed. The microbial DNA will be extracted from different specimens, including breast milk, and maternal and infant stool samples using commercial kits. Quality control and metagenomics sequencing will be employed and analysed by bioinformatics experts to obtain microbiome information including taxonomy and function.

Patient and public involvement

The public representative was involved in the Joint CUHK-NTEC Clinical Research Ethics Committee, whose membership includes lay persons from the community, apart from clinical and nursing staffs. Results of the trial will be disseminated to the community including study participants through presentations in workshops and posters as well as flyers that will be posted in the outpatient clinics for ease of access. Plan for participant involvement in participant recruitment is not considered in this study. This trial has subsequent follow-up schedules. In each visit, participants will be asked for any adverse effects. A diary form is also prepared to mark down the dose taken and any side effects.