Advertisement

Probiotics and the immunological response to infant vaccinations; a double-blind randomized controlled trial

Open AccessPublished:August 09, 2018DOI:https://doi.org/10.1016/j.cmi.2018.07.031

      Abstract

      Objectives

      To examine the effect of a combination of probiotics on the antibody response to pneumococcal and pertussis vaccination in healthy Danish children, aged 8–14 months, at the time of starting day care. Moreover, the cytokine response to lipopolysaccharide of whole blood was assessed.

      Methods

      A total of 290 children were randomly allocated to receive a combination of Bifidobacterium animalis ssp. lactis and Lactobacillus rhamnosus GG daily for a 6-month intervention period, and blood samples were drawn at the start and end of the study. Specific antibody response towards Streptococcus pneumoniae serotypes and Bordetella pertussis toxin, as well as endotoxin-induced interleukin-6 (IL-6) and interferon-γ (IFN-γ) production in blood were analysed by Luminex and ELISA.

      Results

      There was no significant difference between the average individual changes from baseline to end of study in antibody concentrations for S. pneumoniae for both the probiotics (340.4% ± 11.2%) and the placebo group (382.9% ± 10.4%) (p 0.525), nor for B. pertussis toxin in the two groups (probiotics 190.1% ± 12.6% versus placebo 238.8% ± 1.1%, p 0.340). The average individual change in IL-6 concentration was significantly lower in the probiotics versus the placebo group (2.9% ± 10.3% versus 33.7% ± 9.0%, p 0.024), whereas there was no difference in IFN-γ concentration (0.0% ± 0.2% versus –0.2% ± 0.1%, p 0.279).

      Conclusions

      The probiotic intervention did not affect the antibody response against S. pneumoniae and B. pertussis toxin in healthy Danish children.

      Keywords

      Introduction

      Around 90% of all Danish children aged 1–2 years attend day care [

      Statistics Denmark. Færre børn i daginstitutionerne efter skolereformen. Available at: http://dst.dk/site/dst/udgivelser/nyt/getpdf.aspx?cid=19245 [Accessed 28 February 2018].

      ]. These children have an increased risk of respiratory and gastrointestinal infections compared with children in home-care [
      • Kamper-Jørgensen M.
      • Wohlfahrt J.
      • Simonsen J.
      • Grønbaek M.
      • Benn C.S.
      Population-based study of the impact of childcare attendance on hospitalizations for acute respiratory infections.
      ,
      • Lu N.
      • Samuels M.E.
      • Shi L.
      • Baker S.L.
      • Glover S.H.
      • Sanders J.M.
      • et al.
      Child day care risks of common infectious diseases revisited.
      ]. Especially the first 6 months of enrolment into day care have been associated with a high incidence of hospitalizations for acute respiratory infections [
      • Kamper-Jørgensen M.
      • Wohlfahrt J.
      • Simonsen J.
      • Grønbaek M.
      • Benn C.S.
      Population-based study of the impact of childcare attendance on hospitalizations for acute respiratory infections.
      ]. Therefore, strategies to reduce and prevent infections at this stage are of great importance.
      The 13-valent pneumococcal conjugate vaccine (PCV13) was introduced in the Danish Childhood Immunization Programme in 2010 [
      • Harboe Z.B.
      • Dalby T.
      • Weinberger D.M.
      • Benfield T.
      • Mølbak K.
      • Slotved H.C.
      • et al.
      Impact of 13-valent pneumococcal conjugate vaccination in invasive pneumococcal disease incidence and mortality.
      ]. The PCV13 has led to a 71% reduction in invasive pneumococcal disease in Danish children <2 years old (considered as the vaccine effectiveness) [
      • Harboe Z.B.
      • Dalby T.
      • Weinberger D.M.
      • Benfield T.
      • Mølbak K.
      • Slotved H.C.
      • et al.
      Impact of 13-valent pneumococcal conjugate vaccination in invasive pneumococcal disease incidence and mortality.
      ]. Vaccination against Bordetella pertussis is an acellular vaccine of pertussis toxoid [
      • Hviid A.
      Effectiveness of two pertussis vaccines in preterm Danish children.
      ] and has a reported efficacy of >70% [
      • Thierry-Carstensen B.
      • Dalby T.
      • Stevner M.A.
      • Robbins J.B.
      • Schneerson R.
      • Trollfors B.
      • et al.
      Experience with monocomponent acellular pertussis combination vaccines for infants, children, adolescents and adults—a review of safety, immunogenicity, efficacy and effectiveness studies and 15 years of field experience.
      ]. It was suggested that the intestinal microbiota may improve vaccine responses as dysbiosis resulted in systemic inflammation and lower vaccine responses [
      • Huda M.N.
      • Lewis Z.
      • Kalanetra K.M.
      • Rashid M.
      • Ahmad S.M.
      • Raqib R.
      • et al.
      Stool microbiota and vaccine responses of infants.
      ]. Likewise, probiotics may facilitate stimulation of the host immune system. Studies and meta-analyses have addressed the effects of probiotics on vaccine responses in both adults and children, and the results indicate a variation in efficacy of stimulating the immune system depending on the use of probiotic strains, population and vaccination type [
      • Praharaj I.
      • John S.M.
      • Bandyopadhyay R.
      • Kang G.
      Probiotics, antibiotics and the immune responses to vaccines.
      ,
      • Lei W.-T.
      • Shih P.-C.
      • Liu S.-J.
      • Lin C.-H.
      • Yeh T.-L.
      Effect of probiotics and prebiotics on immune response to influenza vaccination in adults: a systematic review and meta-analysis of randomized controlled trials.
      ,
      • Zimmermann P.
      • Curtis N.
      The influence of probiotics on vaccine responses—a systematic review.
      ]. In the studies reported so far about half report a beneficial effect. As the studies, apart from the vaccine regimen, vary with respect to type and number of strains, dose, intervention period, age and group size, no clear conclusions of the effect of probiotics on antibody response to vaccination is available. Similarly, there are no indications of which children may benefit immunologically from a probiotic intervention.
      The ProbiComp study investigated the effect of a daily administration of a probiotic mixture, for 6 months to children from the enrolment in day care at the age of 8–14 months, on absence due to infections [
      • Laursen R.P.
      • Larnkjær A.
      • Ritz C.
      • Hauger H.
      • Michaelsen K.F.
      • Mølgaard C.
      • et al.
      Probiotics and child care absence due to infections: a randomized controlled trial.
      ]. As a secondary outcome, we here report the effects of the probiotics on immunological responses to vaccine exposure to PCV13 and to pertussis toxoid. In addition, as an alternative measure of the probiotic effect on the humoral response, we assessed the capability of cytokine production after ex vivo lipopolysaccharide (LPS) -stimulated whole blood.

      Methods

       Design and study population

      This study was part of the ProbiComp study, a double-blind, placebo-controlled parallel study described in detail by Laursen et al. [
      • Laursen R.P.
      • Larnkjær A.
      • Ritz C.
      • Hauger H.
      • Michaelsen K.F.
      • Mølgaard C.
      • et al.
      Probiotics and child care absence due to infections: a randomized controlled trial.
      ]. In short, 290 Danish children (8–14 months) were randomly assigned to a combination of probiotics Lactobacillus rhamnosus GG (LGG) and Bifidobacterium animalis ssp. lactis (BB-12) at a dose of 2 × 109 CFUs or placebo (maltodextrin) each for a 6-month period. LGG and BB-12 are registered trademarks of Chr. Hansen A/S. Recruitment was carried out for two seasons, from mid-August to mid-December in 2014 and 2015. The children were recruited from the National Danish Civil Registry by the principal investigators of the ProbiComp study [
      • Laursen R.P.
      • Larnkjær A.
      • Ritz C.
      • Hauger H.
      • Michaelsen K.F.
      • Mølgaard C.
      • et al.
      Probiotics and child care absence due to infections: a randomized controlled trial.
      ]. The probiotic and placebo powders were provided in identical sachets and did not differ in smell, taste, or colour. Study personnel and parents were unaware of the nature of the product. The children were assigned to receive either probiotics or a placebo by using block randomization with a block size of eight and unblinding was performed only after completion of data collection and statistical analyses. Children underwent clinical examinations and blood samples were taken at two visits, before the start of day care and entering intervention, and after 6 months, at the end of the intervention. The parents were interviewed upon inclusion about household characteristics and infant history of illness since birth.

       Vaccination data

      In Denmark, pneumococcal conjugated vaccine (PCV13) and vaccination against B. pertussis toxin (PT), are both administered at 3 months, 5 months and 12 months of age. At baseline, children were at the ages 8–13 months and by end of the study their ages were 14–19 months. Vaccination data were also obtained for each participating child to assess the time between last given vaccination and blood sampling time. Vaccination data were obtained from the Danish Vaccination Register. Vaccination coverage of participating children are presented in Table 1.
      Table 1Characteristics of study participants with blood samples from visits 1 and 2 (n = 213)
      ParameterProbiotics group (n = 104)Placebo group (n = 109)p value
      Female, n (%)47 (45.2)53 (48.6)0.616
      Pearson's chi-squared test.
      Siblings, n (%)56 (51.4)53 (48.6)0.446
      Pearson's chi-squared test.
      Breastfeeding prevalence:
       Visit 1, n (%)51 (49)48 (44)0.464
      Pearson's chi-squared test.
       Visit 2, n (%)11 (10.6)8 (7.3)0.407
      Pearson's chi-squared test.
      Indoor pets, n (%)23 (22.1)23 (21.1)0.857
      Pearson's chi-squared test.
      Daily smoking indoors, n (%)1 (0.2)0 (0)0.305
      Pearson's chi-squared test.
      Health status since birth:
       Doctor-diagnosed asthma or allergic rhinitis, n (%)0 (0)0 (0)1.000
       Doctor-diagnosed atopic dermatitis, n (%)10/20 (50)8/21 (38.1)0.443
      Pearson's chi-squared test.
      Age:
       Visit 1 (median months, IQR)10 (9.4–10.4)10.1 (9.6–10.5)0.347
      Welch two sample t-test.
       Start day care (median months, IQR)10.3 (9.9–11.2)10.4 (10–11.2)0.428
      Welch two sample t-test.
       Visit 2 (median months, IQR)16.1 (15.6–16.5)16.0 (15.6–16.6)0.527
      Welch two sample t-test.
      Vaccination:
       PCV13 coverage ≤1 at visit 1, n (%)6 (5.8)9 (8.3)0.490
      Pearson's chi-squared test.
       PCV13 coverage ≥2 at visit 1, n (%)97 (93.3)100 (91.7)0.490
      Pearson's chi-squared test.
       PCV13 coverage ≤2 at visit 2, n (%)16 (15.4)12 (11)0.345
      Pearson's chi-squared test.
       PCV13 coverage ≥3 at visit 2, n (%)88 (87.6)97 (89)0.345
      Pearson's chi-squared test.
       No PCV13 vaccinations1 (1)0 (0)0.305
      Pearson's chi-squared test.
       PT coverage ≤1 at visit 1, n (%)6 (5.8)7 (6.4)0.842
      Pearson's chi-squared test.
       PT coverage ≥2 at visit 1, n (%)98 (94.2)102 (93.6)0.088
      Pearson's chi-squared test.
       PT coverage ≤2 at visit 2, n (%)15 (14.4)12 (11)0.954
      Pearson's chi-squared test.
       PT coverage ≥3 at visit 2, n (%)88 (84.6)97 (89)0.438
      Pearson's chi-squared test.
       No PT vaccinations1 (1)0 (0)0.303
      Pearson's chi-squared test.
       No info visit 1 or visit 22 (1.9)0(0)0.146
      Pearson's chi-squared test.
       PCV13 vaccine before visit 1 (median days, IQR)141 (123.2–159.2)141 (126.8–160)0.858
      Welch two sample t-test.
       PCV13 vaccine before visit 2 (median days, IQR)105 (89–137.2)104 (80–126)0.093
      Welch two sample t-test.
       PT vaccine before visit 1 (median days, IQR)141 (125–160)141.9 (126.8–229)0.509
      Welch two sample t-test.
       PT vaccine before visit 2 (median days, IQR)105 (89–137)108.5 (80–126)0.073
      Welch two sample t-test.
      Compliance:
       Second visit (median % of consumed probiotics, IQR)97 (93–99)97 (94–99)0.987
      Welch two sample t-test.
      IQR, interquartile range; PCV, pneumococcal conjugated vaccine; PT, pertussis toxin; Visit 1, baseline at inclusion; Visit 2, end-point at end of study.
      a Pearson's chi-squared test.
      b Welch two sample t-test.

       Blood sampling and sample preparation

      Peripheral whole blood was drawn at baseline and end of study, and aliquots (approx. 0.5 mL) were subsequently sampled according to the type of analysis. Serum was collected and stored at –80°C until examination by Luminex and ELISA. For LPS stimulation of whole blood, freshly drawn blood was used within 30 min.

       Antibody measurements of S. pneumoniae serotypes and B. pertussis toxin

      Antibody measurements against specific anti-pneumococcal capsular IgG (IgG-PN) were performed for 12 S. pneumoniae serotypes: 1, 3, 4, 5, 6B, 7F, 9V, 14, 18C, 19A, 19F and 23F, using a Luminex-based assay previously described by Kantsø et al. [
      • Kantsø B.
      • Ingdam Halkjær S.
      • Østergaard Thomsen O.
      • Belard E.
      • Gottschalck I.B.
      • Jørgensen C.S.
      • et al.
      Immunosuppressive drugs impairs antibody response of the polysaccharide and conjugated pneumococcal vaccines in patients with Crohn’s disease.
      ]. The geometric mean of the antibody measurements against the 12 serotypes was used for analysis and all ‘out of range’ high values were set to 50 mg/L for all calculations.
      IgG antibodies to B. pertussis toxin (IgG-PT) were measured by an ELISA previously described by Dalby et al. [
      • Dalby T.
      • Seier-Petersen M.
      • Kristiansen M.P.
      • Harboe Z.B.
      • Krogfelt K.A.
      Problem solved: a modified enzyme-linked immunosorbent assay for detection of human antibodies to pertussis toxin eliminates false-positive results occurring at analysis of heat-treated sera.
      ,
      • Dalby T.
      • Petersen J.W.
      • Harboe Z.B.
      • Krogfelt K.A.
      Antibody responses to pertussis toxin display different kinetics after clinical Bordetella pertussis infection than after vaccination with an acellular pertussis vaccine.
      ]. Results were expressed as IU/mL according to the WHO International Standard Pertussis Antiserum National Institute for Biological Standards and Control code 06/140.

       Interleukin-6 and interferon-γ production in LPS-stimulated whole blood

      Freshly drawn whole blood was stimulated essentially as previously described [
      • Lauritzen L.
      • Kjær T.M.R.
      • Fruekilde M.-B.
      • Michaelsen K.F.
      • Frøkiær H.
      Fish oil supplementation of lactating mothers affects cytokine production in 2 1/2-year-old children.
      ]. In brief, duplicates of 50 μL EDTA-treated whole-blood were diluted in 450 μL RPMI-1640 medium, stimulated with 1 mg/L LPS from Escherichia coli O26:B6 (Sigma Aldrich, St Louis, MO, USA), and incubated at 37°C and 5% CO2 for 24 h. Interleukin-6 (IL-6) and interferon-γ (IFN-γ) concentration in the supernatants were measured by sandwich-ELISA using antibody pairs (R&D Systems, Minneapolis, MN, USA). The detection limits for the IL-6 and IFN-γ assays were 147 ng/L and 15 ng/L, respectively, and intra- and inter-assay coefficient of variation were 14% and 8% for IL-6 and 18% and 4% for IFN-γ, respectively.

       Ethical considerations

      The protocol for the study was approved by the Committees on Biomedical Research Ethics for the Capital Region of Denmark (H-4-2014-032). Furthermore, the study was registered at clinicaltrials.gov (identifier NCT02180581 posted 2 July 2014). This study required informed consent from parents and legal guardians of the children. Participation in the study was voluntary and parents could withdraw their consent at any time [
      • Laursen R.P.
      • Larnkjær A.
      • Ritz C.
      • Hauger H.
      • Michaelsen K.F.
      • Mølgaard C.
      • et al.
      Probiotics and child care absence due to infections: a randomized controlled trial.
      ]. Vaccination data were obtained at the Danish Vaccination Register (record number 2015-57-0102).

       Statistics

      Baseline characteristics are shown as the mean (SD) or median (interquartile range (IQR)) for continuous variables and n (%) for categorical variables, and difference between the probiotics and placebo groups was evaluated by Pearson's chi-squared test or Welch two-sample t-test. The effect of probiotics on the change from baseline to end of study was evaluated by using mixed linear models with simultaneous tests for general linear hypothesis. Changes from baseline to end of study and differences between treatment groups at baseline or end of study were tested by selected pairwise comparisons. Due to the logarithmic nature of antibody titre data, the % change from baseline to end of study and % difference between treatment groups were calculated from the estimate given by R using the formula (exp(Estimate) – 1)*100, and the SD was similarly calculated from the standard error given by R using the formula (exp(standard error) – 1)*100. Both % change and % difference and their corresponding SD values for cytokine responses were directly provided by R. Only children with blood samples available for both visits were included in the analyses. All statistical analyses were performed in the software R (version 3.4.3; R Core Team, 2017. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria), using the lme4 package.

      Results

       Study population

      In total, 260 of the 290 children completed the study (Fig. 1, see ref. [
      • Laursen R.P.
      • Larnkjær A.
      • Ritz C.
      • Hauger H.
      • Michaelsen K.F.
      • Mølgaard C.
      • et al.
      Probiotics and child care absence due to infections: a randomized controlled trial.
      ] for more details). It was possible to collect blood from 252 children at intervention start and 231 children at end of study, and 213 children delivered blood samples at both visits. The aliquot volume of collected blood was, however, not always sufficient, limiting the analyses to be made. The numbers of analyses are shown in Fig. 1.
      Fig. 1
      Fig. 1Flowchart of participants and number of blood samples analysed.
      The probiotic and placebo groups had similar distributions regarding sociodemographic factors, duration of breastfeeding, health status since birth, time between sampling and vaccinations, and study compliance (Table 1).

       Antibody response towards S. pneumoniae and B. pertussis vaccination

      We assessed the geometric mean of the antibody concentrations against 12 S. pneumoniae serotypes for a total of 160 children and measured human IgG antibodies against PT in serum samples from 152 children.
      No significant difference was observed in the S. pneumoniae antibody response measured between the group of children given probiotics and placebo, neither at baseline nor end of study (Table 2). The antibody concentrations measured at end of study (probiotics 5.18 mg/L, IQR 2.15–9.26 and placebo 5.62 mg/L, IQR 2.35–11.91) were significantly higher than baseline concentrations (probiotics 1.11 mg/L, IQR 0.58–1.68 and placebo 1.02 mg/L, IQR 0.72–1.81, p ≤ 0.001) (Table 2). However, there was no difference in the change from baseline to end of study in antibody response between the treatment groups (p 0.525) (Table 2, see Supplementary material, Fig. S1). Also, the time (measured in days) since last vaccination at both baseline and end of study significantly affected the concentrations (at baseline a decrease of 0.95%/day; at end of study a decrease of 0.97%/day, p ≤ 0.001) (see Supplementary material, Fig. S2).
      Table 2Antibody response towards vaccination
      Streptococcus pneumoniae (IgG-PNGM mg/L)

      (n = 160)
      Bordetella pertussis toxin (IgG-PT IU/mL)

      (n = 152)
      Baseline median (IQR)End of study median (IQR)Change %p valueBaseline median (IQR)End of study median (IQR)Change %p value
      Probiotic group (n = 74)1.11 (0.58–1.68)5.18 (2.15–9.26)340.4≤0.001
      Mixed linear model with selected pairwise comparisons.
      Probiotic group (n = 70)19.46 (11.09–33.53)57.29 (35.30–101.67)190.1≤0.001
      Mixed linear model with selected pairwise comparisons.
      Placebo group (n = 86)1.02 (0.72–1.81)5.62 (2.35–11.91)382.9≤0.001
      Mixed linear model with selected pairwise comparisons.
      Placebo group (n = 82)17.64 (9.55–30.64)57.87 (32.27–101.99)238.8≤0.001
      Mixed linear model with selected pairwise comparisons.
      Difference %4.6–14.7Difference %13.4–1.2
      p value0.988
      Mixed linear model with selected pairwise comparisons.
      0.749
      Mixed linear model with selected pairwise comparisons.
      0.525
      Mixed linear model with simultaneous tests for general linear hypothesis.
      p value0.727
      Mixed linear model with selected pairwise comparisons.
      1.000
      Mixed linear model with selected pairwise comparisons.
      0.340
      Mixed linear model with simultaneous tests for general linear hypothesis.
      IQR, interquartile range.
      a Mixed linear model with selected pairwise comparisons.
      b Mixed linear model with simultaneous tests for general linear hypothesis.
      Likewise, there was no significant difference in the PT antibody response measured between the group of children given probiotics and placebo, neither at baseline nor end of study (Table 2). The concentrations of PT antibodies were significantly different from baseline to end of study. In the probiotic group, the antibody concentrations measured at end of study (57.29 IU/mL, IQR 35.30–101.67) were significantly higher than baseline concentrations (19.46 IU/mL, IQR 11.09–33.53, p ≤ 0.001) (Table 2). In the placebo group, the antibody concentrations measured at end of study (57.87 IU/mL, IQR 32.27–101.99) were significantly higher than baseline concentrations (17.64 IU/mL, IQR 9.55–30.64, p ≤ 0.001) (Table 2). There was no difference in the change from baseline to end of study in antibody response between the treatment groups (p 0.340) (Table 2, see Supplementary material, Fig. S1). However, the PT antibody concentrations measured at both baseline and end of study depended on the time (measured in days) since last vaccination (at baseline decrease of 0.75%/day and at end of study 0.63%/day, p ≤ 0.001 (see Supplementary material, Fig. S2).

       Whole blood LPS-stimulation-induced IL-6 and IFN-γ responses

      Concentrations of IL-6 changed from baseline (6139 ng/mL, IQR 2840–9625) to end of study (9798 ng/mL, IQR 5547–12 322) in LPS-stimulated whole blood in the placebo group (33.7% ± 9.0%, p ≤ 0.001) (Table 3). No change from baseline (8234 ng/mL, IQR 3078–12 055) to end of study (8278 ng/mL, IQR 3276–9625) was observed in the probiotics group (2.85% ± 10.27%, p 0.993) (Table 3). This resulted in an overall significant difference in the change in LPS-stimulated IL-6 concentration from baseline to end of study between treatment groups (p 0.0238) In contrast, there was no difference in the change of LPS-stimulated IFN-γ production from baseline to end of study between the two groups (p 0.279) (Table 3, see Supplementary material, Fig. S3). Specifically, concentrations of IFN-γ changed from baseline (42.00 ng/mL, IQR 6.05–106.17) to end of study (55.33 ng/mL, IQR 19.33–80.67) in LPS-stimulated whole blood in the placebo group (–0.23% ± 0.14%, p 0.383) (Table 3). No change from baseline (41.88 ng/mL, IQR 6.00–95.00) to end of study (56.00 ng/mL, IQR 15.67–87.33) was observed in the probiotics group (0.01% ± 0.17%, p 1.000) (Table 3).
      Table 3Lipopolysaccharide-induced cytokine responses in whole blood
      Interleukin-6 (ng/mL)

      (n = 118)
      Interferon-γ (ng/mL)

      (n = 96)
      Baseline median (IQR)End of study median (IQR)Change %p valueBaseline median (IQR)End of study median (IQR)Change %p value
      Probiotic group (n = 51)8234 (3078–12055)8278 (3276–9625)2.90.993
      Mixed linear model with selected pairwise comparisons.
      Probiotic group (n = 41)41.88 (6.00–95.00)56.00 (15.67–87.33)0.01.000
      Mixed linear model with selected pairwise comparisons.
      Placebo group (n = 67)6139 (2840–9625)9798 (5547–12322)33.7≤0.001
      Mixed linear model with selected pairwise comparisons.
      Placebo group (n = 55)42.00 (6.05–106.17)55.33 (19.33–80.67)–0.20.383
      Mixed linear model with selected pairwise comparisons.
      Difference %–12.218.7Difference %–0.20.1
      p value0.616
      Mixed linear model with selected pairwise comparisons.
      0.235
      Mixed linear model with selected pairwise comparisons.
      0.0238
      Mixed linear model with simultaneous tests for general linear hypothesis.
      p value0.728
      Mixed linear model with selected pairwise comparisons.
      0.979
      Mixed linear model with selected pairwise comparisons.
      0.279
      Mixed linear model with simultaneous tests for general linear hypothesis.
      IQR, interquartile range.
      a Mixed linear model with selected pairwise comparisons.
      b Mixed linear model with simultaneous tests for general linear hypothesis.

      Discussion

      We have assessed the effect of a daily intake of a combination of probiotic strains, LGG and BB-12, on the response to vaccinations against S. pneumoniae and B. pertussis in healthy children starting in day care as a secondary outcome of the ProbiComp study [
      • Laursen R.P.
      • Larnkjær A.
      • Ritz C.
      • Hauger H.
      • Michaelsen K.F.
      • Mølgaard C.
      • et al.
      Probiotics and child care absence due to infections: a randomized controlled trial.
      ]. It was previously shown that the administered probiotic strains could be detected in faeces and proliferated in the gut, but did not change the gut microbiota community structure [
      • Laursen M.F.
      • Laursen R.P.
      • Larnkjær A.
      • Michaelsen K.F.
      • Bahl M.I.
      • Licht T.R.
      Administration of two probiotic strains during early childhood does not affect the endogenous gut microbiota composition despite probiotic proliferation.
      ]. In this study, we did not find an effect on the concentration of specific antibody response between the probiotic and placebo group towards the two vaccines. Yet, the vaccine response was high (Table 2). Moreover, to study a systemic effect of the probiotics, we also assessed changes in the capability to produce cytokines upon ex vivo LPS stimulation in whole blood and found here significant difference between the IL-6 production in blood from the group receiving probiotics and the placebo group.
      Only a limited number of other studies have assessed the effect of probiotics on the outcome of vaccination in children. To our knowledge, seven other studies have addressed the effect of probiotics on the antibody response against toxoid vaccines and carbohydrate vaccines in children/infants. They vary greatly in number of participants, duration, and dose and strains of administered probiotics. Of these, three studies administered probiotics to infants from birth in a period of 4–12 months [
      • Kukkonen K.
      • Nieminen T.
      • Poussa T.
      • Savilahti E.
      • Kuitunen M.
      Effect of probiotics on vaccine antibody responses in infancy—a randomized placebo-controlled double-blind trial.
      ,
      • Soh S.E.
      • Ong D.Q.R.
      • Gerez I.
      • Zhang X.
      • Chollate P.
      • Shek L.P.
      • et al.
      Effect of probiotic supplementation in the first 6 months of life on specific antibody responses to infant hepatitis B vaccination.
      ,
      • Wu B.-B.
      • Yang Y.
      • Xu X.
      • Wang W.P.
      Effects of Bifidobacterium supplementation on intestinal microbiota composition and the immune response in healthy infants.
      ]. One study assessed the effect of administering a probiotic mixture of around 1.2 × 1010 CFU daily to newborn infants during the first 6 months of life. Antibody titres against diphtheria and tetanus toxoid in the 6-month-old children showed no significant differences between the probiotic and placebo group, but for Haemophilus influenzae type B, concentrations tended to be higher in the probiotic group [
      • Kukkonen K.
      • Nieminen T.
      • Poussa T.
      • Savilahti E.
      • Kuitunen M.
      Effect of probiotics on vaccine antibody responses in infancy—a randomized placebo-controlled double-blind trial.
      ]. The same trend of Haemophilus influenzae type B IgG level was observed in another study [
      • Soh S.E.
      • Ong D.Q.R.
      • Gerez I.
      • Zhang X.
      • Chollate P.
      • Shek L.P.
      • et al.
      Effect of probiotic supplementation in the first 6 months of life on specific antibody responses to infant hepatitis B vaccination.
      ]. A recent study also did not find any differences after probiotic supplementation in diphtheria, tetanus, pertussis, polio and hepatitis B antibody titres [
      • Wu B.-B.
      • Yang Y.
      • Xu X.
      • Wang W.P.
      Effects of Bifidobacterium supplementation on intestinal microbiota composition and the immune response in healthy infants.
      ].
      Four studies initiated the probiotic intervention at a later time-point, between 2 months and 5 years [
      • Matsuda F.
      • Chowdhury M.I.
      • Saha A.
      • Asahara T.
      • Nomoto K.
      • Tarique A.A.
      • et al.
      Evaluation of a probiotics, Bifidobacterium breve BBG-01, for enhancement of immunogenicity of an oral inactivated cholera vaccine and safety: a randomized, double-blind, placebo-controlled trial in Bangladeshi children under 5 years of age.
      ,
      • Pérez N.
      • Iannicelli J.C.
      • Girard-Bosch C.
      • González S.
      • Varea A.
      • Disalvo L.
      • et al.
      Effect of probiotic supplementation on immunoglobulins, isoagglutinins and antibody response in children of low socio-economic status.
      ,
      • West C.E.
      • Gothefors L.
      • Granström M.
      • Käyhty H.
      • Hammarström M.L.
      • Hernell O.
      Effects of feeding probiotics during weaning on infections and antibody responses to diphtheria, tetanus and Hib vaccines.
      ,
      • Youngster I.
      • Kozer E.
      • Lazarovitch Z.
      • Broide E.
      • Goldman M.
      Probiotics and the immunological response to infant vaccinations: a prospective, placebo controlled pilot study.
      ]. Of these studies, the study by Pérez et al. [
      • Pérez N.
      • Iannicelli J.C.
      • Girard-Bosch C.
      • González S.
      • Varea A.
      • Disalvo L.
      • et al.
      Effect of probiotic supplementation on immunoglobulins, isoagglutinins and antibody response in children of low socio-economic status.
      ] is most comparable with ours, as they also addressed the IgG response to pneumococcal vaccination and saw no effect of the probiotic intervention. However, in contrast to our study, the enrolled children varied greatly in age (from 9 months to 10 years), and were from families of low socio-economic status and presumed to have been exposed to a high microbial load. Such factors are expected to affect the study outcome, e.g. by higher variation due to time since last vaccination, or by reducing the effect of the probiotic intervention due to the high microbial load. Even though the children in the present study came from families in Denmark where the majority of parents had >15 years of education, so indicating a ‘high hygienic lifestyle’, no effect of the probiotic intervention was found. Hence, the importance of the microbial stimulation for the strength of the antibody response towards these types of vaccines is questionable. Of note, in our study the probiotic intervention took place during the first months in day care, a time where the infant encounters a high number of microorganisms of high diversity, and so may—with respect to the microbial load—not differ substantially from the Pérez et al. study [
      • Pérez N.
      • Iannicelli J.C.
      • Girard-Bosch C.
      • González S.
      • Varea A.
      • Disalvo L.
      • et al.
      Effect of probiotic supplementation on immunoglobulins, isoagglutinins and antibody response in children of low socio-economic status.
      ]. Taken together, our study is in line with former studies, which found no or few trends towards an increase in antibody levels towards the injected antigens (notably against Haemophilus influenzae type B).
      We also performed another analysis of systemic blood parameters; cytokine production (IL-6 and IFN-γ) upon ex vivo LPS-stimulated whole blood. The production of IL-6 by whole blood reflects the number and response of innate leucocytes, notably monocytes, while IFN-γ is produced by natural killer cells and T helper type 1 cells [
      • Schirmer M.
      • Smeekens S.P.
      • Vlamakis H.
      • Jaeger M.
      • Oosting M.
      • Franzosa E.A.
      • et al.
      Linking the human gut microbiome to inflammatory cytokine production capacity.
      ,
      • Mærkedahl R.B.
      • Frøkiær H.
      • Stenbæk M.G.
      • Nielsen C.B.
      • Lind M.V.
      • Lundtoft C.
      • et al.
      In vivo and ex vivo inflammatory markers of common metabolic phenotypes in humans.
      ]. Here, we found that the change in IL-6 level differed significantly between the probiotic and placebo groups. This may imply that probiotics may affect the capability to respond to a microbial stimulus or LPS. However, as the difference seems to be greatly driven by a lower IL-6 baseline level in the placebo group, we cannot rule out that the difference is due to regression to the mean in the placebo group.
      Number of probiotic bacteria as well as the strain(s) used seems to influence the outcome [
      • Zimmermann P.
      • Curtis N.
      The influence of probiotics on vaccine responses—a systematic review.
      ]. In our study, the dose was 2 × 109 CFU/day, which is comparable to most other infant studies. The combination of LGG and BB-12 has not been used in any of the reported studies, hence we cannot exclude that other strains would have given a different outcome.
      Most of the studies showing an enhancing effect of the antibody titre enrolled individuals above 65 years [
      • Yeh T.-L.
      • Shih P.-C.
      • Liu S.-J.
      • Lin C.H.
      • Liu J.M.
      • Lei W.T.
      • et al.
      The influence of prebiotic or probiotic supplementation on antibody titers after influenza vaccination: a systematic review and meta-analysis of randomized controlled trials.
      ], indicating that a positive effect of probiotics may depend on the immune status of the participants, as it is well-known that the immune system deteriorates with age. As the effects of probiotics on vaccination are mostly seen in studies involving elderly individuals, this points towards the fact that vaccines today are so efficient that an effect of probiotics is merely on the general status of the immune system. Hence, we suggest that immunocompromised individuals and elderly will benefit from probiotics by improved protection upon vaccination.
      A strength in the current study is that blood samples were drawn at enrolment as well as at the termination of the intervention, allowing us to baseline adjust end-point values. A limitation in the study is that we did not test against viral vaccines (rubella, mumps, measles), which might have given another outcome. However, in other studies children did not show markedly different outcome of probiotic intervention on the viral vaccines compared with the toxoid and carbohydrate vaccines [
      • Zimmermann P.
      • Curtis N.
      The influence of probiotics on vaccine responses—a systematic review.
      ,
      • Wu B.-B.
      • Yang Y.
      • Xu X.
      • Wang W.P.
      Effects of Bifidobacterium supplementation on intestinal microbiota composition and the immune response in healthy infants.
      ]. We studied the effects of probiotics in healthy Danish children assumed to have a well-developed immune system. A group of immunocompromised children might have given other results.
      The prevailing theories regarding the mechanism(s) by which probiotics exert their putative effects are still debated. In general, the probiotic bacteria are believed to, like other bacteria, stimulate the innate immune system but may also affect the polarization of the adaptive immune system [
      • Levy M.
      • Kolodziejczyk A.A.
      • Thaiss C.A.
      • Elinav E.
      Dysbiosis and the immune system.
      ]. We have recently shown that probiotics administered to newborn mice with a depleted microbiota increased granulopoiesis (unpublished data), a mechanism that may explain a positive effect of probiotics on influenza-vaccinated elderly. This mechanism may also explain the significant difference in ex vivo IL-6 production in LPS-stimulated whole blood, as a high concentration of neutrophils in blood may affect the concentration of monocytes, the main IL-6 producers in blood. This is, however, purely speculative because we cannot rule out the possibility of regression to the mean in the placebo group. Probiotics may also aid the stimulation of defence mechanisms in the epithelium, notably production of mucins and anti-microbial peptides. This mechanism may explain why more pronounced effects are found in studies of probiotics and mucosal vaccines [
      • Isolauri E.
      • Joensuu J.
      • Suomalainen H.
      • Luomala M.
      • Vesikari T.
      Improved immunogenicity of oral D x RRV reassortant rotavirus vaccine by Lactobacillus casei GG.
      ,
      • Paineau D.
      • Carcano D.
      • Leyer G.
      • Darquy S.
      • Alyanakian M.A.
      • Simoneau G.
      • et al.
      Effects of seven potential probiotic strains on specific immune responses in healthy adults: a double-blind, randomized, controlled trial.
      ].
      In conclusion, we did not find an effect of probiotic intervention on the antibody titres to pneumococcal and pertussis toxoid vaccination in 14- to 18-month-old children after their first months in day care. This may reflect how efficient the vaccines are and suggests that only immunocompromised individuals and elderly may benefit from probiotics to strengthen the outcome of vaccination.

      Transparency declaration

      The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be constructed as a potential conflict of interest.

      Funding

      The ProbiComp study was funded by Innovation Fund Denmark (grant number 0603-00579B ) to Kim F. Michaelsen, Department of Nutrition, Exercise and Sports, Faculty of Science, University of Copenhagen, Denmark.

      Acknowledgements

      The authors wish to thank the children and their parents for participating in this study. Second, they thank Anni Mehlsen and Inge B. Rasmussen for technical assistance in blood sampling and ELISA analyses, and our colleagues at Statens Serum Institut, Hans-Christian Slotved for conversations and guidance concerning pneumococcal infections and vaccinations, and Sine Fjeldhøj for obtaining vaccination data for all the children enrolled in the study.

      Access to data

      Rikke Pilmann Laursen and Kim F. Michaelsen, Department of Nutrition, Exercise and Sports, Faculty of Science, University of Copenhagen, Denmark.

      Author contributions

      CAS and EF have contributed equally to this article and share first authorship. CAS and EF contributed to analysis and interpretation of data, drafting of the article. CSJ contributed to analysis of antibody concentrations. HF contributed to analysis and interpretation of cytokine data. RPL coordinated and conducted the data collection. AL and CM contributed to design of the ProbiComp study. CR contributed to statistical data analysis. KFM contributed to study design and obtained approvals. KAK and HF contributed to study design, planning analyses and drafting of the article. All authors critically reviewed the manuscript, approved the final draft and take responsibility for the integrity and the accuracy of this research and the interpretation hereof.

      Appendix A. Supplementary data

      Fig. S1
      Fig. S1Antibody response towards 13-valent pneumococcal conjugate vaccine (PCV13) or Bordetella pertussis toxin (PT) at baseline to end of study (PCV13: placebo, n = 86; probiotics, n = 103. PT: placebo, n = 55; probiotics, n = 41).
      Fig. S2
      Fig. S2Impact of time (measured in days) since last vaccination on antibody responses in all children included.
      Fig. S3
      Fig. S3Cytokine concentrations upon ex vivo lipopolysaccharide (LPS) stimulation of whole blood at baseline to the end of study (interleukin: placebo, n = 67; probiotics, n = 51. interferon-γ: placebo, n = 81; probiotics, n = 103).

      References

      1. Statistics Denmark. Færre børn i daginstitutionerne efter skolereformen. Available at: http://dst.dk/site/dst/udgivelser/nyt/getpdf.aspx?cid=19245 [Accessed 28 February 2018].

        • Kamper-Jørgensen M.
        • Wohlfahrt J.
        • Simonsen J.
        • Grønbaek M.
        • Benn C.S.
        Population-based study of the impact of childcare attendance on hospitalizations for acute respiratory infections.
        Pediatrics. 2006; 118: 1439-1446
        • Lu N.
        • Samuels M.E.
        • Shi L.
        • Baker S.L.
        • Glover S.H.
        • Sanders J.M.
        • et al.
        Child day care risks of common infectious diseases revisited.
        Child Care Heal Dev. 2004; 30: 361-368
        • Harboe Z.B.
        • Dalby T.
        • Weinberger D.M.
        • Benfield T.
        • Mølbak K.
        • Slotved H.C.
        • et al.
        Impact of 13-valent pneumococcal conjugate vaccination in invasive pneumococcal disease incidence and mortality.
        Clin Infect Dis. 2014; 59: 1066-1073
        • Hviid A.
        Effectiveness of two pertussis vaccines in preterm Danish children.
        Vaccine. 2009; 27: 3035-3038
        • Thierry-Carstensen B.
        • Dalby T.
        • Stevner M.A.
        • Robbins J.B.
        • Schneerson R.
        • Trollfors B.
        • et al.
        Experience with monocomponent acellular pertussis combination vaccines for infants, children, adolescents and adults—a review of safety, immunogenicity, efficacy and effectiveness studies and 15 years of field experience.
        Vaccine. 2013; 31: 5178-5191
        • Huda M.N.
        • Lewis Z.
        • Kalanetra K.M.
        • Rashid M.
        • Ahmad S.M.
        • Raqib R.
        • et al.
        Stool microbiota and vaccine responses of infants.
        Pediatrics. 2014; 134: e362-e372
        • Praharaj I.
        • John S.M.
        • Bandyopadhyay R.
        • Kang G.
        Probiotics, antibiotics and the immune responses to vaccines.
        Phil Trans R Soc B. 2015; 370: 1-12
        • Lei W.-T.
        • Shih P.-C.
        • Liu S.-J.
        • Lin C.-H.
        • Yeh T.-L.
        Effect of probiotics and prebiotics on immune response to influenza vaccination in adults: a systematic review and meta-analysis of randomized controlled trials.
        Nutrients. 2017; 9: 1-17
        • Zimmermann P.
        • Curtis N.
        The influence of probiotics on vaccine responses—a systematic review.
        Vaccine. 2018; 36: 207-213
        • Laursen R.P.
        • Larnkjær A.
        • Ritz C.
        • Hauger H.
        • Michaelsen K.F.
        • Mølgaard C.
        • et al.
        Probiotics and child care absence due to infections: a randomized controlled trial.
        Pediatrics. 2017; 140: 1-11
        • Kantsø B.
        • Ingdam Halkjær S.
        • Østergaard Thomsen O.
        • Belard E.
        • Gottschalck I.B.
        • Jørgensen C.S.
        • et al.
        Immunosuppressive drugs impairs antibody response of the polysaccharide and conjugated pneumococcal vaccines in patients with Crohn’s disease.
        Vaccine. 2015; 33: 5464-5469
        • Dalby T.
        • Seier-Petersen M.
        • Kristiansen M.P.
        • Harboe Z.B.
        • Krogfelt K.A.
        Problem solved: a modified enzyme-linked immunosorbent assay for detection of human antibodies to pertussis toxin eliminates false-positive results occurring at analysis of heat-treated sera.
        Diagn Microbiol Infect Dis. 2009; 63: 354-360
        • Dalby T.
        • Petersen J.W.
        • Harboe Z.B.
        • Krogfelt K.A.
        Antibody responses to pertussis toxin display different kinetics after clinical Bordetella pertussis infection than after vaccination with an acellular pertussis vaccine.
        J Med Microbiol. 2010; 59: 1029-1036
        • Lauritzen L.
        • Kjær T.M.R.
        • Fruekilde M.-B.
        • Michaelsen K.F.
        • Frøkiær H.
        Fish oil supplementation of lactating mothers affects cytokine production in 2 1/2-year-old children.
        Lipids. 2005; 40: 669-676
        • Laursen M.F.
        • Laursen R.P.
        • Larnkjær A.
        • Michaelsen K.F.
        • Bahl M.I.
        • Licht T.R.
        Administration of two probiotic strains during early childhood does not affect the endogenous gut microbiota composition despite probiotic proliferation.
        BMC Microbiol. 2017; 17: 1-9
        • Kukkonen K.
        • Nieminen T.
        • Poussa T.
        • Savilahti E.
        • Kuitunen M.
        Effect of probiotics on vaccine antibody responses in infancy—a randomized placebo-controlled double-blind trial.
        Pediatr Allergy Immunol. 2006; 17: 416-421
        • Soh S.E.
        • Ong D.Q.R.
        • Gerez I.
        • Zhang X.
        • Chollate P.
        • Shek L.P.
        • et al.
        Effect of probiotic supplementation in the first 6 months of life on specific antibody responses to infant hepatitis B vaccination.
        Vaccine. 2010; 28: 2577-2579
        • Wu B.-B.
        • Yang Y.
        • Xu X.
        • Wang W.P.
        Effects of Bifidobacterium supplementation on intestinal microbiota composition and the immune response in healthy infants.
        World J Pediatr. 2016; 12: 177-182
        • Matsuda F.
        • Chowdhury M.I.
        • Saha A.
        • Asahara T.
        • Nomoto K.
        • Tarique A.A.
        • et al.
        Evaluation of a probiotics, Bifidobacterium breve BBG-01, for enhancement of immunogenicity of an oral inactivated cholera vaccine and safety: a randomized, double-blind, placebo-controlled trial in Bangladeshi children under 5 years of age.
        Vaccine. 2011; 29: 1855-1858
        • Pérez N.
        • Iannicelli J.C.
        • Girard-Bosch C.
        • González S.
        • Varea A.
        • Disalvo L.
        • et al.
        Effect of probiotic supplementation on immunoglobulins, isoagglutinins and antibody response in children of low socio-economic status.
        Eur J Nutr. 2010; 49: 173-179
        • West C.E.
        • Gothefors L.
        • Granström M.
        • Käyhty H.
        • Hammarström M.L.
        • Hernell O.
        Effects of feeding probiotics during weaning on infections and antibody responses to diphtheria, tetanus and Hib vaccines.
        Pediatr Allergy Immunol. 2008; 19: 53-60
        • Youngster I.
        • Kozer E.
        • Lazarovitch Z.
        • Broide E.
        • Goldman M.
        Probiotics and the immunological response to infant vaccinations: a prospective, placebo controlled pilot study.
        Arch Dis Child. 2011; 96: 345-349
        • Schirmer M.
        • Smeekens S.P.
        • Vlamakis H.
        • Jaeger M.
        • Oosting M.
        • Franzosa E.A.
        • et al.
        Linking the human gut microbiome to inflammatory cytokine production capacity.
        Cell. 2016; 167: 1125-1136
        • Mærkedahl R.B.
        • Frøkiær H.
        • Stenbæk M.G.
        • Nielsen C.B.
        • Lind M.V.
        • Lundtoft C.
        • et al.
        In vivo and ex vivo inflammatory markers of common metabolic phenotypes in humans.
        Metab Syndr Relat Disord. 2018; 16: 29-39
        • Yeh T.-L.
        • Shih P.-C.
        • Liu S.-J.
        • Lin C.H.
        • Liu J.M.
        • Lei W.T.
        • et al.
        The influence of prebiotic or probiotic supplementation on antibody titers after influenza vaccination: a systematic review and meta-analysis of randomized controlled trials.
        Drug Des Devel Ther. 2018; 12: 217-230
        • Levy M.
        • Kolodziejczyk A.A.
        • Thaiss C.A.
        • Elinav E.
        Dysbiosis and the immune system.
        Nat Rev Immunol. 2017; 17: 219-232
        • Isolauri E.
        • Joensuu J.
        • Suomalainen H.
        • Luomala M.
        • Vesikari T.
        Improved immunogenicity of oral D x RRV reassortant rotavirus vaccine by Lactobacillus casei GG.
        Vaccine. 1995; 13: 310-312
        • Paineau D.
        • Carcano D.
        • Leyer G.
        • Darquy S.
        • Alyanakian M.A.
        • Simoneau G.
        • et al.
        Effects of seven potential probiotic strains on specific immune responses in healthy adults: a double-blind, randomized, controlled trial.
        FEMS Immunol Med Microbiol. 2008; 53: 107-113