Safety and tolerability of Bifidobacterium longum subspecies infantis EVC001 supplementation in healthy term breastfed infants: a phase I clinical trial

Abstract

Background: Historically, bifidobacteria were the dominant intestinal bacteria in breastfed infants. Still abundant in infants in developing nations, levels of intestinal bifidobacteria are low among infants in developed nations. Recent studies have described an intimate relationship between human milk and a specific subspecies of Bifidobacterium, B. longum subsp. infantis (B. infantis), yet supplementation of breastfed, healthy, term infants with this organism, has not been reported. The IMPRINT Study, a Phase I clinical trial, was initiated to determine the safety and tolerability of supplementing breastfed infants with B. infantis (EVC001).

Methods: Eighty mother-infant dyads were enrolled in either lactation support plus B. infantis supplementation (BiLS) or lactation support alone (LS). Starting with Day 7 postnatal, BiLS infants were fed 1.8-2.8 × 1010 CFU B. infantis EVC001 daily in breast milk for 21 days. Mothers collected fecal samples, filled out health questionnaires, and kept daily logs about their infants' feeding and gastrointestinal symptoms from birth until Day 61 postnatal. Safety and tolerability were determined from maternal reports.

Results: There were no differences in the mean gestational age at birth, weight 1 and 2 months postnatal, and breast milk intake between groups. The mean Log10 change in fecal Bifidobacterium from Day 6 to Day 28 was higher (p = 0.0002) for BiLS (6.6 ± 2.8 SD) than for LS infants (3.5 ± 3.5 SD). Daily stool number was higher (p < 0.005) for LS and lower (p < 0.05) for BiLS infants during supplementation than at Baseline. During supplementation, watery stools decreased and soft stools increased by 36% over baseline in BiLS infants (p < 0.05) with no significant changes in stool consistency for the LS infants. None of the safety and tolerability endpoints, including flatulence, bloody stool, body temperature, ratings of gastrointestinal symptoms, use of antibiotics or gas-relieving medications, infant colic, jaundice, number of illnesses, sick doctor visits, or diagnoses of eczema were different for the groups at any point.

Conclusions: The B. infantis EVC001 supplement was safely consumed and well-tolerated. Stools were fewer and better formed in infants in the BiLS group compared with LS group. Adverse events were those expected in healthy infants and not different between groups.

Trial registration: ClinicalTrials.gov NCT02457338 . Registered May 27, 2015.

Keywords: Bifidobacterium longum subspecies infantis; Breast milk; Gut microbiome; Human milk oligosaccharides; Infant; Probiotics; Supplementation; Tolerability.

Figures

Fig. 1

Consort diagram. Consort diagram describing the number of participating mothers who were screened, randomized into the intervention groups, screened-failed post-randomization, and withdrew throughout the study period

Fig. 2

Number of infant stools per day. Mean ± SD of reported number of daily infant stools for the LS (red dot plot) and BiLS (blue dot plot) groups during the Baseline, Intervention, and Post-intervention periods. n = 34 for each group during the Baseline and Intervention periods, and n = 33 for the LS, and n = 34 for the BiLS groups during the Post-intervention period. Different superscripts represent significant differences within and between interventions. There was a significant time effect (p < 0.01), time*trt interaction (p < 0.0005), and intervention effect (p < 0.0005). Based on multiple comparison post hoc analysis with Bonferroni corrections, compared with baseline the mean number of stools increased (bp < 0.0005) during the intervention period for the LS group and decreased (cp < 0.05) for the BiLS group. During the Post-intervention period, the mean number of stools returned to Baseline levels for the LS group and decreased from the Intervention period for the BiLS group (dp < 0.0005)

Fig. 3

Infant stool consistency. a Mean ± SD of the proportion in reported infant stool consistency for the LS (red dot plot) and BiLS (blue dot plot) groups during the Baseline, Intervention, and Post-intervention periods. n = 34 for each group during the Baseline and Intervention periods, n = 33 for the LS group, and n = 34 for the BiLS group during the Post-intervention period. b Mean ± SD of the change in the percentage of reported infant stool consistency for the LS (red dot plot) and BiLS (blue dot plot) groups for difference between Intervention and Baseline (Intervention – Baseline), and Post-intervention and Intervention (Post-intervention – Intervention) periods. n = 34 for each group for Intervention – Baseline, and n = 33 for the LS and n = 34 for the BiLS groups for Post-intervention – Intervention. *p < 0.05

Fig. 4

Infant tolerability scores. Mean ± SD of reported tolerability scores (red dot plot) and BiLS (blue dot plot) groups during the Baseline, Intervention, and Post-intervention periods. n = 34 for each group during the Baseline and Intervention periods, and n = 33 for the LS and n = 34 for the BiLS groups during the Post-intervention period. a Infant irritability, (b) infant upset after spit-ups, and (c) infant discomfort in passing gas or stool