Breast milk jaundice (BMJ), benign unconjugated hyperbilirubinemia associated with breast-feeding, is a common cause of prolonged jaundice in otherwise healthy breast-fed infants born at term (1). BMJ presents in the first or second week of life, and can persist for as long as 12 weeks before spontaneous resolution. The incidence of BMJ in the exclusively breast-fed infant during the first 2 to 3 weeks of life has been reported at 36% (2). Despite the fact that prolonged jaundice caused by breast milk is a common occurrence in the neonatal period, a large number of tests are required to rule out pathological causes (3).
Although many theories have been proposed to explain BMJ, the jaundice of breast-fed infants is commonly of undetermined etiology. Presently, the increased intestinal absorption of bilirubin, and the resultant increase in its enterohepatic circulation, appears to be one of the most convincing mechanisms to explain the neonatal jaundice associated with breast-feeding (1,4,5). An important consideration related to intestinal bilirubin absorption is the establishment of a population of intestinal bacteria that converts bilirubin glucuronides to various urobilinoids, and therefore reduces the availability of bilirubin for intestinal reabsorption (6,7).
Recent studies have demonstrated that human milk, far from being a sterile fluid, constitutes an excellent and continuous source of commensal bacteria for the infant gut. Staphylococcus, Streptococcus, Bifidobacterium, and Lactobacillus are the most commonly determined bacterial spp (species) in human milk (8–12). Molecular analysis shows that these bacteria are metabolically active in the human gut, increasing the production of functional metabolites such as butyrate, which is the main energy source for colonocytes and plays a key role in the modulation of intestinal function (13).
We hypothesized that bacteria in breast milk may play a role in reducing the occurrence of BMJ because of their influence on the gut microflora. The present study aims to investigate the effects of the bacterial content of breast milk and the infant's feces on the occurrence of BMJ.
The present study was approved by the Human Ethics Committee of the Dokuz Eylul University Faculty of Medicine. The participating women received verbal and written information about the aim and structure of the study, and they were asked for written consent to participate.
A total of 600 infant-mother pairs were screened at the Newborn Outpatient Clinic, and 60 infant-mother pairs were enrolled consecutively to the study according to the following selection criteria: infants who were fed exclusively by breast-feeding; healthy women without present or past underlying conditions; normal, full-term pregnancy; and absence of infant and/or maternal perinatal problems, including mastitis. None of the mothers enrolled in the present study had received a probiotic treatment during pregnancy or after birth, and none of them had taken antibiotics therapy after birth. Also, the infants did not receive antibiotics, probiotics, probiotic-supplemented products, or probiotic- and prebiotic-supplemented formulas before their enrollment and thereafter. The participants provided samples of breast milk and infant feces between the 14th and 28th postnatal days; the mother-infant pairs failed to provide proper breast milk and infant feces samples were not included in the study.
Thirty infants who developed prolonged jaundice and were considered BMJ were enrolled to the study. BMJ has been defined as jaundice beginning after 5 to 7 days of life, and peaking around the 10th day or later of life. BMJ has been diagnosed when other causes of prolonged jaundice are ruled out, and serum bilirubin levels return to normal before 3 months in fully breast-fed infants (2). Infants were excluded who had known risk factors, such as blood group incompatibilities, positive Coombs test, glucose-6-phosphate dehydrogenase deficiencies, any laboratory evidence of hemolytic disease as evidenced by anemia, reticulocytosis or abnormality of the blood smear, and perinatal factors associated with an increased risk of hyperbilirubinemia, including maternal diabetes mellitus, polycythemia, cephalohematoma, asphyxia, hypothermia, intracranial hemorrhage, or perinatal infection. The control group was composed of 30 healthy infants who were born in our hospital and had no clinical jaundice during their follow-up period.
Sample collection date and time were recorded, as were the infants’ birth dates, gestational ages, birth weights, sex, mothers’ demographic, race, and anthropometrical characteristics (age, parity, body mass index), and route of delivery (cesarean section versus vaginal).
Milk and Feces Specimen Collection
Breast milk samples were collected during the third and the fourth postpartum week. For this purpose, nipples and mammary areola were cleaned with soap and sterile water, and then chlorhexidine was applied. The breast milk sample was collected in a sterile tube after manual expression using sterile gloves. The first drops (approximately 0.5 mL) were discarded. Milk samples were taken bilaterally and then 5 mL of these specimens were put into sterile tubes. Fecal samples of the infants were also put into sterile tubes. All samples were placed on ice, immediately sent to the laboratory, and then stored at −70°C until polymerase chain reaction (PCR) analysis.
Bacterial DNA Isolation From Feces
Fecal samples of each infant were homogenized, and 150 mg of homogenized feces were obtained from each sample. A ZR Fecal DNA MiniPrep (Zymo Research Corp, Irvine, CA) was used for the isolation of bacterial DNA from the homogenized stool according to the manufacturer's instructions (14).
Bacterial DNA Isolation From Breast Milk
After homogenization of breast milk samples, 300 μL of homogenized breast milk samples were centrifuged at 3000g for 5 minutes. High Pure PCR Template Preparation Kit (Roche Applied Science, Penzberg, Germany) was used for the isolation of bacterial DNA from lysozyme-treated breast milk according to the manufacturer's instructions.
Primers and Universal Probe Library Probes
Eleven different genus-specific primer sets were used in the present study. A Multiple Sequence Alignment Web tool, CLUSTALW2 (http://www.ebi.ac.uk), was used to identify the homologous regions of bacterial strains for each bacteria. Primer sets were designed using the Primer-BLAST program (NCBI, GenBank, BLAST), and Universal Probe Library probes were selected by using ProbeFinder version 2.45, a Web-based software tool (http://www.roche-applied-science.com).
Real-time PCR reaction for bacterial DNA extracted from breast milk and fecal samples, in duplicates, was carried out in a total volume of 20 μL with a reaction mixture. Fifteen different bacterial species were screened for each sample. Bacterial species that were screened in the infant's feces and the breast milk samples were the following: Bifidobacterium spp (B bifidum, B adolescentis, B longum), Lactobacillus spp (L gasseri, L rhamnosus, L fermentum, L plantarum), Staphylococcus spp (S epidermidis, S hominis, S aureus), Streptococcus spp (S salivarius, S mitis), Clostridium spp (C perfringens, C difficile), and Bacteroides spp (B fragilis).
The results were analyzed by using the “Abs Quant/2nd Derivative Max” method with the LightCycler 480 II (Roche) analysis program. The results were expressed as “Crossing Point value” (cycle number in a log-linear region). Calculated Crossing Point values are inversely correlated with microorganism concentrations (15).
All analyses were performed using computer software SPSS for Windows release 15.01 (SPSS Inc, Chicago, IL). The bacterial counts in feces and in breast milk were non-normally distributed, and thus nonparametric statistical methods were applied. The Mann-Whitney test was used to test differences between the 2 groups. Unless otherwise specified, the results are expressed as mean (±standard error) and median (upper-lower quartiles). A probability level of <0.05 was considered to be statistically significant. The Pearson Correlation Index was used to calculate the relation between milk's and feces’ microbiological concentrations and peak bilirubin levels. The association between frequencies was tested using Fisher exact test. An estimated sample size of 30 was determined to be necessary to detect a relative difference of 0.50 between groups in terms of breast milk microbial colonization with 80% power and a 2-tailed α level of 0.05.
The Hospital Ethical Committee for Human Research of the participating hospital approved the research protocol.
Thirty infants with prolonged jaundice and 30 healthy infants for the control group were included in the study. Fecal DNA isolation could not be managed in 1 sample from the control group because of technical reasons. The 2 groups were similar in terms of maternal, fetal, and neonatal demographic characteristics. The mode of delivery, which may potentially affect our results, was similar in the 2 groups. The total bilirubin levels at the time of sample collection for the study and control groups were 14.0 ± 2.7 and 3.8 ± 1.1, respectively. There was no significant difference between the 2 groups in terms of specimen collection day (17.4 ± 2.6 vs 18.1 ± 2.0).
Of all studied bacteria, Bifidobacterium spp was the most common microorganism in breast milk samples. The most frequently detected Bifidobacterium spp were B bifidum (86%), B adolescentis (62%), and B longum (32%). The proportion of positive breast milk samples for different Bifidobacterium and Lactobacillus spp in the 2 groups is shown in Figure 1. As illustrated, the presence of these microorganisms in breast milk was higher in the control group in comparison with the BMJ group, but was not statistically significant.
When we compared the 2 groups in terms of the concentration of these microorganisms in breast milk samples, B adolescentis and B bifidum concentrations were found to be higher in the control group than in the jaundiced group (P < 0.01). Although the breast milk concentration of Lactobacillus spp seemed to be higher in the control group than in the jaundiced group, there was no statistical significance (Table 1).
When the microbial content of fecal samples was compared, concentrations of B adolescentis, B longum, and B bifidum spp were found to be significantly higher in the control group than in the jaundiced group (P < 0.01). There was no significant difference between the 2 groups in terms of fecal Lactobacillus spp concentrations. No significant difference between the 2 groups was found in terms of clostridial organisms and other bacterial species.
Correlations between the feces and breast milk bacterial concentrations were evaluated for the microorganisms showing a significant concentration difference between the 2 groups. Only the concentrations of B bifidum showed a positive correlation between the feces and breast milk samples (Fig. 2). Also, the correlation between the concentration of the microorganisms in the milk and feces samples and serum bilirubin levels was evaluated. The concentrations of B adolescentis, B bifidum, and B longum in feces samples were negatively correlated with the serum bilirubin levels (r = −0.88, P < 0.001; r = −0.77, P < 0.001; and r = −0.43, P = 0.005, respectively). Although a negative correlation was detected between the breast milk's B bifidum concentration and the serum bilirubin levels (Fig. 3), breast milk's B adolescentis and B longum concentrations showed no correlation with the bilirubin levels.
A wide array of hypotheses has been proposed in an attempt to understand the mechanism of BMJ. It was believed that an increased absorption of bilirubin played a key role in BMJ (1,2). To our knowledge, the present study is the first to suggest that bacteria in breast milk may play a role in reducing the occurrence of BMJ, influencing the gut microflora.
The influence of intestinal microflora on serum bilirubin levels was first shown directly in hyperbilirubinemic Gunn rats. Oral administration of a wide-spectrum antibiotic therapy resulted in the disappearance of fecal urobilinoids and, simultaneously, in a dramatic increase in serum bilirubin levels. Furthermore, intestinal colonization with C perfringens led to the reappearance of fecal urobilinoids with a partial decrease in serum bilirubin levels (16). Only 4 bacterial strains capable of bilirubin conversion have been isolated with certainty so far: C ramosum, C perfringens, C difficile, and Bacteroides fragilis(17). In the present study, effects of feces’ Clostridium spp concentrations on prolonged jaundice were not observed.
Bifidobacterium spp comprise the predominant intestinal bacteria concentration in full-term, breast-fed infants at as early as 3 to 6 days of age (12,18,19). More recently, the Bifidobacterium predominance in the intestinal microbiota of breast-fed infants has been linked to the direct transfer of maternal bifidobacteria to newborns in breast milk (11,12). In the present study, 86% of breast milk samples were found to contain Bifidobacterium spp and concentrations of Bifidobacterium in breast milk and feces were positively correlated. In a previous study using similar methods, all of the breast milk samples were found to contain Bifidobacterium spp but no correlation was found between the total count of Bifidobacterium in breast milk and the infants’ feces (20); however, another study covering a large number of various countries found Bifidobacterium spp in breast milk samples with varying frequencies (0%–100%), implying that the microbial colonization of breast milk seems to be highly dependent on the bacteriological status of the society (21). This may be related to the other factors affecting the growth of bifidogenic bacteria in the breast milk, such as oligosaccharides, or the colonization of the breast milk by other environmental bacteria.
Prebiotic levels in maternal milk and their relation with fecal probiotic and bilirubin levels could not be evaluated in our study. But it can be speculated that prebiotics may be protective against hyperbilirubinemia by affecting intestinal motility and intestinal microbial flora. In a prospective randomized controlled study including formula-fed healthy term newborns, the addition of prebiotics to a standard infant diet resulted in lower bilirubin levels (22). The association of prebiotic supplementation with lower bilirubin levels supports the role of microbiota in BMJ, since prebiotics modified intestinal microbiota.
Although the biological properties of probiotic microorganisms have been well known for many years, the data related to the influence of intestinal probiotic bacteria on serum bilirubin levels in infants are scarce. How enteric probiotic microorganisms could influence the occurrence of jaundice remains in question. As already suggested, probiotic bacteria increase the fecal moisture, frequency, and volume of stool. Moreover, they contribute to the development of an intestinal barrier through the formation of mucin (23). They also regulate the human epithelial tight junction and protect the epithelial integrity (24). Our results suggest that intestinal probiotic bacteria may be protective against hyperbilirubinemia possibly through the reduction in bilirubin absorption.
The main potential limitations of the present study were the use of a semi-quantitative method for the detection of bacterial concentrations by real-time PCR, and an absence of confirmation by culture-dependent methods, and the inability to demonstrate possible action mechanisms of these microorganisms. It should be noted that there is no evidence that BMJ is actually harmful, and only the infants that approach potentially toxic levels need to be treated (2). Mild to moderate levels of unconjugated hyperbilirubinemia may be protective to the newborn infant by providing antioxidant effects, which are otherwise absent in the newborn period (25,26).
In conclusion, preliminary results presented here show that breast milk microbial content and the composition of enteric microbiota may play a role not only in the regulation of gut immunology and nonimmunologic defense mechanisms, but also in BMJ. Particularly high concentrations of Bifidobacterium spp in maternal milk and infants’ feces seem to be protective against BMJ. Because BMJ is not actually harmful, and mild to moderate levels of unconjugated hyperbilirubinemia are protective against oxidative stress, the biological plausibility of the potential effects of these bacteria in terms of reducing bilirubin levels is questionable.
The authors are grateful to Cankut Cubuk and Ceren Senkal for technical assistance.
1. Gartner LM. Breastfeeding and jaundice. J Perinatol 2001; 21 1:25–29.
2. Preer GL, Philipp BL. Understanding and managing breast milk jaundice. Arch Dis Child Fetal Neonatal Ed 2011; 96:461–466.
3. Hannam S, McDonnell M, Rennie JM. Investigation of prolonged neonatal jaundice. Acta Paediatr 2000; 89:694–697.
4. Gourley GR, Gourley MF, Arend R, et al. The effect of saccharolactone on rat intestinal absorption of bilirubin in the presence of human breast milk. Pediatr Res 1989; 25:234–238.
5. Alonso EM, Whitington PF, Whitington SH, et al. Enterohepatic circulation of nonconjugated bilirubin in rats fed with human milk. J Pediatr 1991; 118:425–430.
6. Gustafsson BE, Lanke LS. Bilirubin and urobilins in germfree, ex-germfree, and conventional rats. J Exp Med 1960; 112:975–981.
7. Saxerholt H, Midtvedt T, Gustafsson BE. Deconjugation of bilirubin conjugates and urobilin formation by conventionalized germ-free rats. Scand J Clin Lab Invest 1984; 44:573–577.
8. Lara-Villoslada F, Olivares M, Sierra S, et al. Beneficial effects of probiotic bacteria isolated from breast milk. Br J Nutr 2007; 98 1:96–100.
9. Collado MC, Delgado S, Maldonado A, et al. Assessment of the bacterial diversity of breast milk of healthy women by quantitative real-time PCR. Lett Appl Microbiol 2009; 48:523–528.
10. Martin R, Jimenez E, Heilig H, et al. Isolation of bifidobacteria from breast milk and assessment of the bifidobacterial population by PCR-denaturing gradient gel electrophoresis and quantitative real-time PCR. Appl Environ Microbiol 2009; 75:965–969.
11. Gueimonde M, Laitinen K, Salminen S, et al. Breast milk: a source of bifidobacteria for infant gut development and maturation? Neonatology 2007; 92:64–66.
12. Martin V, Maldonado-Barragan A, Moles L, et al. Sharing of bacterial strains between breast milk and infant feces. J Hum Lact 2012; 28:36–44.
13. Olivares M, Diaz-Ropero MP, Gomez N, et al. The consumption of two new probiotic strains, Lactobacillus gasseri CECT 5714 and Lactobacillus coryniformis CECT 5711, boosts the immune system of healthy humans. Int Microbiol 2006; 9:47–52.
14. Yoshikawa H, Dogruman-Ai F, Turk S, et al. Evaluation of DNA extraction kits for molecular diagnosis of human Blastocystis subtypes from fecal samples. Parasitol Res 2011; 109:1045–1050.
15. Abdulamir A, Yoke TS, Nordin N, et al. Detection and quantification of probiotic bacteria using optimized DNA extraction, traditional and real-time PCR methods in complex microbial communities. Af J Biotechnol 2010; 10:1481–1492.
16. Vitek L, Zelenka J, Zadinova M, et al. The impact of intestinal microflora on serum bilirubin levels. J Hepatol 2005; 42:238–243.
17. Vitek L, Kotal P, Jirsa M, et al. Intestinal colonization leading to fecal urobilinoid excretion may play a role in the pathogenesis of neonatal jaundice. J Pediatr Gastroenterol Nutr 2000; 30:294–298.
18. Turroni F, Peano C, Pass DA, et al. Diversity of bifidobacteria within the infant gut microbiota. PloS ONE 2012; 7 5:36957.
19. Tsuji H, Oozeer R, Matsuda K, et al. Molecular monitoring of the development of intestinal microbiota in Japanese infants. Benef Microbes 2012; 3:113–125.
20. Gronlund MM, Gueimonde M, Laitinen K, et al. Maternal breast-milk and intestinal bifidobacteria guide the compositional development of the Bifidobacterium microbiota in infants at risk of allergic disease. Clin Exp Allergy 2007; 37:1764–1772.
21. Sinkiewicz G NoE. Occurrence of Lactobacillus reuteri, Lactobacilli and Bifidobacteria in human breast milk. Pediatr Res 2005;58:415.
22. Bisceglia M, Indrio F, Riezzo G, et al. The effect of prebiotics in the management of neonatal hyperbilirubinaemia. Acta Paediatr 2009; 98:1579–1581.
23. Mack DR, Michail S, Wei S, et al. Probiotics inhibit enteropathogenic E. coli adherence in vitro by inducing intestinal mucin gene expression. Am J Physiol 1999; 276:941–950.
24. Karczewski J, Troost FJ, Konings I, et al. Regulation of human epithelial tight junction proteins by Lactobacillus plantarum in vivo and protective effects on the epithelial barrier. Am J Physiol Gastrointest Liver Physiol 2010; 298:851–859.
25. Gopinathan V, Miller NJ, Milner AD, et al. Bilirubin and ascorbate antioxidant activity in neonatal plasma. FEBS Lett 1994; 349:197–200.
26. Shekeeb Shahab M, Kumar P, Sharma N, et al. Evaluation of oxidant and antioxidant status in term neonates: a plausible protective role of bilirubin. Mol Cell Biochem 2008; 317:51–59.