Probiotics are defined as a live microbial supplement, which beneficially affects the host animal by improving its microbial balance (1). Lactobacillus rhamnosus strain GG, commonly known as Lactobacillus GG (LGG), has been shown to possess the characteristics of an effective probiotic and has been tested in a variety of situations in different age groups (1–3). Reported beneficial effects of LGG include reduction in duration and severity of acute viral diarrhea, antibiotic-associated diarrhea, and diarrhea associated with HIV infection. It may also play a positive role in irritable bowel syndrome and inflammatory bowel disease in adults (1–3). It appears to modify the immune response and the bacterial gut colonization pattern (1–3).
Low birth weight is an important risk factor in neonatal mortality and various morbidities, including sepsis (4). Many low-birth-weight infants require NICU care and receive antibiotics and intravenous fluids for prolonged periods. The bacterial flora of these infants is different from that of normal infants and is characterized by a delay in gut colonization and a paucity of bacterial species, in contrast to the rich mixed flora of multiple aerobic and anaerobic strains seen in full-term infants (5–7). There may also be bacterial overgrowth by a limited number of species, often by Gram-negative bacilli, some of which may be resistant to antibiotics (5–7). Previous treatment with antibiotics and the need to be nursed in an incubator have been associated with a lower rate of early colonization with lactobacilli (8). It has been suggested that bacterial adherence to intestinal cells, influenced by the underlying microbial ecology, can have a role in the pathogenesis of NEC (9). Co-infection with specific Gram-positive bacterial species blocked adherence of Gram-negative organisms to Caco-2 cells in vitro and also blocked E. coli-induced gut injury in a weanling rabbit ileal loop model (9).
Lactobacillus GG modifies the intestinal microflora in adults and children and has been reported to be of clinical benefit in a variety of gastrointestinal and non-gastrointestinal conditions (1). However, experience in preterm newborns has been limited, and reports of efficacy have been variable (10–12). The present study was therefore designed to assess the colonizing capability of LGG and its effects on gut microbial ecology in infants less than 2000 g birth weight.
MATERIALS AND METHODS
Population and Design
This randomized, prospective trial was performed at the All India Institute of Medical Science in New Delhi between December 1999 and June 2000. The study protocol was approved by the Institutional Ethics Committee. Informed consent from parents was obtained before enrollment of the infants. All live-born infants with birth weights less than 2000 g were eligible. Infants were stratified into two groups of birth weight: less than 1500 g and 1500 to 1999 g. Randomization was designed to yield a treated:control ratio of 2:1. Infants with major congenital malformations, hydrops fetalis, those infants not fed orally for more than 72 hours after birth, and those infants discharged within the first 3 days of life were excluded.
Lactobacillus GG (Trade name “Culturelle,” containing 109 lyophilized cells per unit; Valio, Ltd., U.S.A.) was administered orally mixed with expressed breast milk feedings twice a day, once the infant was on oral feeds. Control babies were given nonsupplemented breast milk feeds. Expressed breast milk was the preferred type of nutrition but was occasionally supplemented with formula when milk supply from the mother was insufficient. In the infants who weighed less than 1500 g, LGG treatment was initiated on day 2 or 3 of life and continued for 21 days (range 19.5 to 22.5 days). Stool samples were collected before starting treatment or at the time of enrollment in the control group (baseline), and again at 7, 14, and 21 days after enrollment. In the infants weighing between 1500 and 1999 gm, LGG was initiated on days 1 to 3 of life and continued for 8 days (range 8 to 9 days). Stool samples were taken before starting treatment or at the time of enrollment in the control group and again on days 4 and 8 after enrollment. Since most of these infants were discharged early, probiotic therapy could not be continued beyond an 8-day period. Data on maternal and neonatal variables including any adverse events related to LGG administration were collected. Standard NICU policies were followed for medical and/or surgical treatment of the infants.
Stool samples were collected under sterile conditions and immediately transported to the bacteriology laboratory situated in the same building for processing. The investigator responsible for microbiologic processing of the samples was blinded to assigned group. A 100-mg sample of stool was dissolved in 1 ml of normal saline to prepare a uniform suspension. Using this suspension, ten-fold serial dilutions were prepared; 0.1 ml of each dilution were plated on sheep blood agar and on MacConkey agar for aerobic culture and on brain heart infusion agar for anaerobic culture. After an incubation period of 24 to 48 hours, colony counts were performed for each colony type. The organisms were identified by conventional biochemical and rapid automated tests (API 20 E test, API staph/ API 20 E strept) (Biomerieux Inc., Durham, NC) for aerobic organisms, and the RapID ANA II System (Innovative Diagnostic System LP, Norcross, GA) for anaerobic organisms. Colonization was defined as the isolation of any LGG colonies in any stool specimen by anaerobic or microaerophilic culture.
Colony counts were expressed as log10 counts. Mean log CFU counts were used for statistical comparisons. (A colony count of zero was treated as zero for calculation purposes). A χ2 test or Fisher exact test (where appropriate) were used for comparison of categorical variables. Other variables were compared using a Wilcoxon Rank-sign test or Friedman's repeated measures ANOVA on ranks, with Student-Newman-Kuells or Dunnett's post-hoc tests, as appropriate.
Baseline characteristics of treated, control, and excluded infants, stratified by birth weight are given in Table 1. Of the infants weighing less than 1500 gm, 24 were enrolled in the treatment group and 15 were enrolled in the control group. Of the infants weighing between 1500 and 1999 g, 23 were enrolled in the treatment group and 9 were enrolled in the control group. Treated and control babies did not differ significantly, with the sole exception of the exclusive use of breast milk, which was significantly greater in 1500 to 1999 g control group compared with the weight matched treated group.
Overall, colonization with LGG occurred in 17/47 treated infants versus 0/24 control infants (P = 0.005). No side effects were observed in babies either fed with or colonized by LGG.
Infants weighing less than 1500 gm
Lactobacillus GG colonization occurred in 4% (1/24) of the treatment group by day 14, increasing to 21% (5/24) on day 21. Colonization at any time occurred in 5/24 (21%) in the treatment group; no colonization occurred in the control group at any time (P = 0.14).
There was a consistent increase in the total number of bacterial species from baseline through day 21 in the infants treated with LGG (Fig. 1), with total species number on days 7, 14 and 21 all being significantly higher than baseline; there was no significant increase in species number noted in control infants over time. This difference was mainly the result of increases in Gram (+) species and anaerobic spp. other than LGG. Control infants of <1500 did not have increased numbers of Gram (+) species or anaerobes after 21 days. On day 21, treated infants had a higher number of bacterial species compared to controls. Prior antibiotic therapy in the 7 days preceding the day-21 stool collection adversely affected colonization. Of the 14 LGG-treated babies on antibiotics none were colonized, whereas 5 of the 10 babies not on prior antibiotic therapy became colonized with LGG (P < 0.01). Antenatal steroid therapy did not appear to have an affect on ability to colonize.
Mean log CFU of Gram (−) bacteria did not change in treated infants weighing less than 1500 g from day 0 through day 21, unlike the situation in control infants where there were significantly increased mean log CFU of Gram (−) species on all days compared with baseline (although this may reflect the lower mean log CFU in the controls on day 0) (Table 2). Gram (+) mean log CFU showed a significant increase in the LGG-treated group between day 0 (3.5 ± 0.9) and day 21 (6.1 ± 0.9) (P < 0.05).
Infants 1500 to 1999 gm
In the treatment group, LGG colonization occurred in 4/23 (17%) by day 4 of life, increasing to 11/23 (47%) by day 8. Ten of the 11 infants colonized on day 8 had not received antimicrobial therapy during the prior 7 days (the lone exception was an infant on amphotericin B); however, a definitive statement regarding antibiotic effect on ability to colonize cannot be made in this cohort, since 10 of the 12 non-colonized infants were also not on antibiotics.
In preterm infants weighing 1500 to 1999 gm, LGG treatment also resulted in an increased species number on day 8 versus day 0 (Fig. 2), but the overall effect was not different from control infants, especially since the slight difference in total species between the two groups was mainly the result of the presence of LGG in about half of the treated infants. There were no significant differences in mean log CFU counts in Gram (+) or Gram (−) species in either control or LGG-treated infants (Table 3).
The consumption of probiotic agents such as lactic acid bacteria first occurred when humans began using fermented foodstuffs. After observing the arrested putrefaction of milk products by lactic acid bacteria, Metchnikoff first proposed the use of probiotics to prolong human life in 1907 (13). Fuller has defined a probiotic as a live microbial feed supplement, which beneficially affects the host animal by improving its microbial balance (14). It is generally agreed that a probiotic must be able to resist the extremes of pH during gastrointestinal transit and be able to colonize, although not necessarily proliferate in the gut, to exert its potential beneficial effects (1).
Lactobacillus GG was reported in 1985 by Gorbach and Goldin as a natural strain (isolated from humans) having the properties of a candidate probiotic (1). LGG has been studied in a variety of clinical settings in different age groups and shown to be beneficial in some circumstances. A recent systematic review (15) of published randomized trials indicated that probiotics reduce the risk of diarrhea lasting more than 3 days, with only LGG showing a consistent effect. The prenatal administration of LGG to mothers with a family history of atopic disease followed by LGG administration to their infants for 6 months after birth resulted in a decrease in the occurrence of atopic eczema by approximately one-half compared with a placebo group (16).
Previous studies of LGG treatment in preterm neonates have reported a successful colonization rate of 80% to 90% (11–12). In contrast, in the current study colonization by LGG was no better than 50% in premature infants of 1500 to 1999 g birth weight and less than 25% in the less-than 1500 g cohort. In another study involving a similar population, our group found that Lactobacillus acidophilus had a colonization rate of more than 60% compared with L. sporogenes (0%) in very low-birth-weight babies (17). Reuman et al. (9) noted an 86% colonization rate with Lactobacillus acidophilus. Bifidobacterium breve has been demonstrated to have 73% colonization in preterm VLBW babies (18). Bennet et al. (19) showed that oral bifidobacteria and lactobacilli could be cultured from the feces of term infants for several days after such treatment was discontinued. In the context of the ability of probiotic strains to change the gut flora, Mautone et al. (20) noted a reduction in Gram (−) flora and a percentage increase in Gram (+) bacilli after feeding a mix of Bifidobacterium bifidum and L acidophilus to a group of 77 neonates.
Colonization with a probiotic is dependent on the interplay of multiple factors in the intestinal milieu. First and foremost is the tolerance to the acid and alkali environments that the bacteria encounter. Once bacteria pass these barriers, luminal immunoglobulins, substrate, prebiotics, and concurrent antibiotic treatment can all affect colonization. In the case of newborns, lactose is probably the primary substrate in the colon along with other sugars and by-products of protein and lipid digestion. Bacterial fermentation of sugars can further change the internal environment in the intestine. Even if the probiotic strain has potential to survive stomach acid and duodenal/jejunal alkalinity and has good adherence to intestinal cells, it may not survive and multiply in the absence of certain prebiotic factors in the colon. In addition, use of antenatal steroids and the type of enteral feeding also influence colonization. In this study, LGG colonization occurred only in babies who were not receiving antibiotics during the study. Although resistant to drugs such as vancomycin (due to the presence of an atypical peptidoglycan cell wall), Lactobacillus strains are sensitive to a variety of commonly used antibiotics; thus, antibiotic use is likely to play a major role in colonization by these organisms. Indeed, examination of the in vitro antibiotic sensitivity pattern of LGG reveals sensitivity to a broad spectrum of antibiotics, such as ampicillin, gentamicin, amikacin, kanamycin, ciprofloxacin, Augmentin, netilmicin, ceftazidime, and penicillin. Thus, the failure of LGG to colonize most premature infants in our study can be explained in part by the common usage of some of the above antibiotics. From a practical standpoint, since most of these low-birth-weight infants receive antibiotics early in life, colonization by probiotic agents (including LGG) may be difficult and may require prolonged therapy (beyond the period of antibiotic use).
Antenatal steroids accelerate the maturation of the gastrointestinal tract (21) and thus might lead to the expression of specific receptors needed for attachment of bacilli. Although the underlying mechanisms remain to be explained, several studies have also observed a reduction in necrotizing enterocolitis after maternal steroid administration (22–23). Although we did not observe any effect of antenatal steroids on colonization, other host-derived factors might have played roles in the colonizing ability of our LGG strain. The lack of colonization in half of the small babies (< 1500 g) not receiving antibiotics suggests that there are other host and bacterial factors involved; indeed, LGG may be an inherently poor colonizer of the newborn intestine.
The observed inverse relationship between days on antibiotics and overall bacterial colonization noted in another previous study by our group suggests that reduction of routine antibiotic exposure and decreased length of therapy for selected nosocomial infections may promote fecal microbial diversity (6). Studies demonstrating delayed overall colonization of the premature neonate's gut and eventual colonization by a paucity of organisms (6,17,24) (in both the United States and India) have led to the speculation that the lack of species diversity in the immature intestine may lead to overgrowth of Gram (−) bacilli and subsequent adherence by and translocation of these strains resulting, in an increased susceptibility to necrotizing enterocolitis and sepsis. Indeed, a number of recent studies have suggested a benefit of probiotic therapy in preventing neonatal necrotizing enterocolitis (25–26).
In conclusion, although LGG shows a moderate degree of colonization in larger premature babies (birth weight 1500 to 1999 gm), the smaller babies weighing less than 1500 g exhibited very poor colonization. It is intriguing to note that although there was no change in the Gram (−) reservoir, as reported by Mautone et al. (20), LGG treatment was associated with an appreciable increase in the number and mean log count of Gram (+) species compared with controls. Induction of such species diversity in spite of poor colonization may have beneficial effects in the immature intestine (9). Starting with a sterile gut, the neonatal intestine undergoes rapid exposure and acquisition of normal flora and reaches a steady state of multiple species amounting to a total of 1013 CFU bacteria in an adult (27). This steady healthy state allows various species (normal flora) to co-exist in harmony without one becoming dominant or replacing others (28). Until such a state is reached (during the growing neonatal period), with extremely limited number of species and total low CFU count, the intestine is prone to colonization by new strains including Gram (−) bacteria and other harmful species (9,29). Increase in the Gram (+) CFU count as a result of LGG therapy would be expected to help the neonatal gut attain such a balanced ecosystem in a shorter time frame and prevent the deleterious effects of other bacterial strains on the developing intestine. In the context of these multifactorial effects in the highly complex intestinal milieu, there is a need for further examination of other modifiers of colonization such as prebiotics (oligosaccharides) and antibiotic exposure that have an impact on the final microflora of the infant intestine. In spite of the recent surge in clinical investigations and reports in the field of prebiotics and probiotics, there remains a need for identification of better candidate strains (with higher colonizing ability) and administration regimens for treating and preventing gastrointestinal conditions in the neonatal population, a group that may benefit the most from such therapy.
Funded by NIH Fogarty Grant TW-00601 (PP). The authors thank Jon Vanderhoof, M.D., and Conagra Foods Inc., Omaha, Nebraska for supplying the LGG preparation.
1. Gorbach SL. Probiotics and gastrointestinal health. Am J Gastroenterol 2000; 95(suppl.): S2–S4.
2. Vanderhoof J. Functional significance of the bowel microflora in gastrointestinal health: Proceedings of a roundtable discussion. Am J Gastroenterol 2000; 95(suppl.): S1.
3. Metchnikoff E. The prolongation of life. Optimistic studies. William Heinemann, London 1907.
4. Aylward GP, Pfeiffer SI, Wright A, Verhulst SJ. Outcome studies of LBW infants published in last decade: A meta-analysis. J Pediatr 1989; 115:515–20.
5. Long SS, Swenson RM. Development of anaerobic fecal flora in healthy newborn infants. J Pediatr 1977; 91:298–301.
6. Gewolb IH, Schwalbe RS, Taciak VL, Harrison TS, Panigrahi. Stool microflora in extremely low birth weight infants. Arch Dis Child Fetal Neonatal Ed 1999; 80:F167–73.
7. Sakata H, Yoshioka H, Fujita K. Development of the intestinal flora in very low birth weight infants compared to normal full term infants. Eur J Pediatr 1985; 144:186–90.
8. Hall MA, Cole CB, Smith SL, Fuller R, Rolles CJ. Factors influencing the presence of fecal lactobacilli in early infancy. Arch Dis Child 1990; 65:185–8.
9. Panigrahi P, Gupta S, Gewolb IH, Morris JG. Occurrence of necrotizing enterocolitis may be dependent on patterns of bacterial adherence and intestinal colonization: Studies in Caco-2 tissue culture and weanling rabbit models. Pediatr Res 1994; 36:115–21.
10. Reuman PD, Duckworth DH, Smith KL, Kagan R, Bucciarelli RL, Ayoub EM. Lack of effect of Lactobacillus on gastrointestinal bacterial colonization in premature infants. Pediatr Infect Dis 1986; 5:663–68.
11. Gronlund MM, Lehtonen OP, Kero P, Saxelin M, Salminen S. Lactobacillus GG supplementation does not reduce fecal colonization of Klebsiella
oxytoca in preterm infants. (Erratum). Acta Pediatr 1997; 86:785–86.
12. Miller MR, Bacon C, Smith SL, Walker V, Hall MA. Enteral feedings of premature infants with Lactobacillus GG. Arch Dis Child 1993; 69:483–87.
13. Metchnikoff, E., The prolongation of life. Optimistic studies. 1907, London: Butterworth-Heinemann.
14. Fuller R. Probiotics in man and animals: a review. J Appl Bacteriol 1989; 66:365–78.
15. Szajeeweska H, Mrukowicz JZ. Probiotics in the treatment and prevention of acute infectious diarrhea in infants and children: a systematic review of published randomized, double–blind, placebo-controlled trials. J Pediatr Gastroenterol Nutr 2001; 33Suppl 2:S17–S25.
16. Kalliomaki M, Salminen S, Arvilommi H, Kero P, Isolauri E. Probiotics in primary prevention of atopic disease: a placebo-controlled trial. Lancet 2001; 357:1076–9.
17. Sharma N, Chaudhry R, Bhushan S, Deorari A, Paul VK, Singh M, Gewolb IH, Panigrahi P: Effect of continuous probiotic treatment on stool microbial patterns in very low birth weight infants. Pediatr Res 2000; 47:168A (abstract).
18. Kitajima H, Sumida Y, Janaka R, Yuki N, Jakayama H, Fujimara M. Early administration of Bifidobacterium breve in preterm infants: Randomized control trial. Arch Dis Child Fetal Neonatal ed. 1997; 76: F101–7.
19. Bennet R, Nord CE, Zetterstrom R. Transient colonization of the gut of newborn infants by orally administered bifidobacteria and lactobacilli. Acta Pediatr 1992; 81:784–7.
20. Mautone A, Monno R, Palmieri P, Brandonisio O. Colonizzazione indotta di Bifidobacterium bifidum e di Lactobacillus acidophilus in neonati ospedalizzati. Minerva Pediatr 1982; 34:245–50.
21. Calado TJ, Mandarim CA, Coutinho VB, Robalinho TI, Countinho HB. Morphometry of the microvilli of the enterocyte in the fetus of Dialphis albiventris before and after treatment with hydrocortisone hemisuccinate. Ital. J Anat Embryol 1994; 99:171–180.
22. Quablan HS, Malkawi HY, Hiasat MS, Hindawi IM, Al-Tanami MI, Abu-Khait SA, Al-Maaitah JF. The effect of antenatal corticosteroid therapy on pregnancies complicated by premature rupture of membranes. Clin Exp Obstet Gynecol 2001; 28:183–86.
23. Smith LM, Quereshi N, Chao CR. Effect of single and multiple courses of antenatal glucocorticoids in preterm newborns less than 30 weeks' gestation. J Matern Fetal Med 2000; 9:131–35.
24. Gupta S, Morris JG, Panigrahi P, Nataro JP, Glass RI, Gewolb IH. Endemic necrotizing enterocolitis. Lack of association with a specific infectious agent. Pediatr Infect Dis J 1994; 13:728–34.
25. Hoyos AB. Reduced incidence of necrotizing enterocolitis associated with enteral administration of Lactobacillus acidophilus and Bifidobacterium infantis to neonates in an intensive care unit. Int J Infect Dis 1999; 3:197–202.
26. Rubaltelli FF, Biadaioli R, Biadaioli DC. Probiotics feeding prevents necrotizing enterocolitis in preterm infants; a prospective double-blind study. Pediatr Res, 2000. 47:346A (abstract).
27. Freter R. Control mechanisms of the large intestinal microflora and its influenceon the host. Acta Gastroenterol Latinoam 1989; 19:197–217.
28. Freter R., Brickner H, Temme SJ. An understanding of the colonization resistance of mammalian intestine requires mathematical analysis. Microecology and Therapy 1986; 16:147–55.
29. Conway P. Lactobacilli: fact and fiction, in The regulatory and protective role of normal flora, R Grun, T Midvedt, and E. Norin editors. 1988. Stockton Press. 263–81.