Fecal flora measurements of breastfed infants using an integrated transport and culturing system

Sir, It is generally recognized that bacterial colonization of the intestine with indigenous flora is an important component in development of the intestinal mucosal barrier function in the human infant (1). There is accumulating evidence that the intestinal flora plays an important role in the postnatal development of the immune system (2) and promotes the absorption of nutrients in the colon (3). As a consequence, the artificial manipulation of the intestinal ecosystem by including preor probiotics in feeding formulas to generate a predominantly bifidobacterial flora even in bottle-fed infants has been attempted (4). Detailed knowledge of the microbial ecology of the human colon is essential to obtain reliable information on the effects of preor probiotics and on the influence of diet on the intestinal microflora. Technical difficulties are still a major barrier in studying the gut flora from stool samples and may represent a limiting factor in studies on the effect of early diet on the establishment of the intestinal bacterial ecosystem in the newborn infant. We therefore studied an integrated system of stool sampling, transporting and culturing in order to investigate the number of bifidobacteria and lactobacilli in stool samples of infants of up to 3 mo of age. Particular attention has been paid to the possible influence of storage time. Fifteen exclusively breastfed term infants were enrolled in the study. The Ethics Committee of the University Hospital approved the protocol of the study and informed parental consent was obtained for each infant before enrolment in the study. The infants were examined on the first day of full breastfeeding, i.e. 120 ml kg 1 d 1 (examination day 1; mean age: 5.3 2.4 d) and were re-examined after a 6 to 8-wk period of breastfeeding (examination day 2). On each examination day, the stool samples were collected in the hospital. Two samples were taken for evaluation of a possible influence by a circadian rhythm. When more than two stools were available for each infant, the first and second samples were collected at intervals of 12 h, the 24-h sampling period starting at 06.00 h. To account for data not normally distributed, the data on the microflora as well as stool frequency and consistency were described as medians and interquartile ranges (IQR, 25th–75th percentiles). The influence of the duration of feeding or the influence of duration of storage on these parameters was investigated using nonparametric tests. A linear regression analysis was performed to investigate the relationship between the two different samples from the same examination day. All tests were performed on an alpha-level of 5%. Pvalues 0.05 were considered as significant. The software used was StatView 5.0 (SAS Institute Inc.). On the second examination day, one stool of the first 5 infants was divided into four portions. Each portion was separately transferred to the transport medium, homogenized and stored at 80°C. The samples were analysed after different storage times. The first portion was analysed within 1 mo, the second after 3 mo, the third after 5 mo and the fourth portion after 8 mo. A portion of a fresh faecal sample (0.2 g) was