Secondary Logo

Journal Logo


Neonatal Necrotizing Enterocolitis: Possible Role of Probiotic Supplementation

Caplan, Michael S.; Jilling, Tamas

Author Information
Journal of Pediatric Gastroenterology and Nutrition: March 2000 - Volume 30 - Issue - p S18-S22
  • Free

Probiotics, or anaerobic bacterial supplementation, have been used for years (over the counter and all over the world) to treat gastrointestinal infections and related ailments. Nonetheless, the effect of these interesting organisms on diseases of intestinal inflammation has not been well delineated. Intestinal inflammation results when an imbalance occurs in the normal physiology of intestinal antigens (primarily bacteria), the mucosal barrier, the intestinal immune system, local and systemic chemical mediators of inflammation, and the microvascular endothelium (1,2). Probiotic supplementation has been shown to modulate several of these processes, leading to prevention or amelioration of intestinal disease in several experimental models (3–6). In this report, we focus on the effect of probiotics on neonatal necrotizing enterocolitis (NEC), a specific intestinal inflammatory disease that primarily afflicts premature neonates.


Neonatal NEC is the most common gastrointestinal emergency in premature infants and is characterized by nonspecific symptoms and signs of abdominal distention, bilious emesis, bloody stools, lethargy, apnea and bradycardia and, occasionally, the specific findings of abdominal tenderness and discoloration (7,8). The diagnosis is confirmed either by the radiologic hallmark of pneumatosis intestinalis, exploratory laparotomy, or autopsy. The disease occurs most commonly in previously fed (>90%), premature (>95%) infants and rarely if ever is diagnosed after discharge from the neonatal intensive care unit.

In the United States, NEC is diagnosed in 11% of premature infants born weighing less than 1500 g (9). Within the United States, there is significant variability among neonatal intensive care units in the reported incidence figures. Furthermore, the incidence of NEC is variable around the world. For example, a recent report from Hong Kong of premature infants weighing less than 1500 g showed NEC diagnosed in 28%, with a reduction to 13% with preventive oral vancomycin therapy (10). In Argentina, in similar birthweight infants, NEC was reported in 14%, with only 7% after early dexamethasone therapy (11). Seven percent of small premature infants in Austria had NEC, whereas no cases were found after enteral immunoglobulin (Ig)A/IgG prophylaxis (12). Of interest, the Japanese report incidence figures as low as 1.5% in infants weighing less than 1500 g (13). Despite these differences in NEC incidence, no accepted hypothesis accounts for this interesting variability.

The mortality rate associated with NEC in most reports ranges between 20% and 30%. An additional 30% require surgical intervention but survive, and 30% to 40% do well with medical management alone. Few complications result from NEC, but strictures with recurrent bowel obstruction are diagnosed in as many as 25% in some reports. Most survivors of NEC do well long term; the neurodevelopmental outcome is unaffected by the presence of this disease (14).


Despite extensive research during the past 30 years, the pathophysiology of NEC remains poorly understood. The current multifactorial theory postulates that the risk factors of prematurity, bacterial colonization, formula feeding, and intestinal ischemia–hypoxia result in the final common pathway of intestinal necrosis (15).

Gastrointestinal host defense is markedly impaired in premature infants (16). It is known that premature animals and humans have an immature mucosal barrier and abnormal mucosal hormone and enzyme function. In addition, young animals have altered mucosal immune function, abnormal intestinal microvascular autoregulation, and inefficient peristalsis. Nonetheless, the specific host defense deficiencies that place the premature infant at risk for NEC remain unknown.

The role of intestinal ischemia in NEC has been studied using animal models and epidemiologic analyses in premature infants (17–19). In animals, complete intestinal ischemia after clamping of the superior mesenteric artery results in only subtle histologic abnormalities, but after reperfusion, severe intestinal necrosis results. In babies, studies have shown that a patent ductus arteriosus (with reduced diastolic blood flow) and the treatment of the ductus with indomethacin (decreased intestinal blood flow through inhibition of cyclooxygenase) both increase the risk for NEC (20). Furthermore, the presence of an umbilical artery catheter or cocaine exposure (both thought to compromise intestinal perfusion) increased the risk for NEC in some studies but not in others. Although the specific mechanisms are unclear, altered intestinal blood flow places the neonate at risk for NEC.

The initiation of feeding is an important risk factor for NEC, and studies have shown a reduction in the disease with breast milk feedings compared with neonatal formula supplementation (21). Nonetheless, the precise mechanisms responsible for this breast milk protection (or formula-induced stress) remain unclear. Although differences in neonatal feeding practices include rates of volume increase, concentration and strength of formula, nasogastric versus nasojejunal approach, continuous versus bolus feeding, and timing of feeding initiation, no clear data specify alterations in risk for NEC (22).

The presence of bacteria colonizing the intestinal lumen appears to be a prerequisite for the development of NEC; no cases have been described in utero before normal gut colonization. Although epidemics of NEC related to a specific bacterial pathogen have been described, these events are unusual, because the disease typically occurs endemically unrelated to host microbiology. The gut bacterial flora of the premature intensive care patient differs from that of the full-term, nursing infant (23). The patient in a hospital has fewer bacterial species and reduced colonization of anaerobic flora, suggesting that overgrowth of specific pathogenic bacteria in certain instances may contribute to the initiation of NEC.


The key risk factors of NEC result in the final common pathway of intestinal necrosis through the activation of the inflammatory cascade. Experimental evidence has shown that concentrations of chemical mediators—platelet-activating factor (PAF), interleukin (IL)-1, IL-6, tumor necrosis factor (TNF), nitric oxide (NO), and endothelin—are elevated in circulation or intestinal homogenate in neonates with NEC and in various models of intestinal injury (15). Furthermore, in animal models, many of these mediators cause bowel necrosis after intravenous administration (24,25). Drugs or antagonists that block the production and/or effects of these compounds inhibit or ameliorate the initiation of NEC (26,27). Finally, NEC is a systemic illness in many cases, with similar signs and symptoms of septic shock attributable to activation of the inflammatory cascade.

Platelet-activating factor (PAF) is a potent phospholipid mediator that appears to play a key role in the pathophysiology of NEC (28). This compound is synthesized after the activation of phospholipase A2-II (PLA2), and is degraded under the influence of the enzyme PAF-acetylhydrolase (PAF-AH) (29,30). It exerts its' effects through the activation of the G-protein–coupled PAF receptor that is present on most cells but is ubiquitous on intestinal epithelium (31).

Intravenous administration of PAF to adult rats results in ischemic bowel necrosis similar to the disease of neonatal NEC. Furthermore, in many models of intestinal injury (e.g., lipopolysaccharide [LPS]-induced, hypoxia-induced, TNF-and LPS-induced, ischemia–reperfusion), concentrations of PAF in intestinal homogenate are elevated, and pretreatment with PAF receptor antagonists reduce the incidence and severity of disease (25,26,32). Human neonatal studies have shown elevated plasma PAF levels and stool PAF concentrations in infants with NEC compared with age-matched control subjects (28). We have studied a newborn rat model of NEC similar to the human disease and found that both formula feeding and asphyxia exposure are necessary for the initiation of NEC (33). Using this approach, we have shown that asphyxia and formula feeding together increase intestinal PLA2-II mRNA and enzyme activity, and PAF-receptor mRNA in neonatal animals. Furthermore, treatment with PAF receptor antagonists reduces the incidence of intestinal injury and death in this model (34). Similarly, animals treated with the PAF-AH enzyme have a marked reduction in NEC incidence (35). The results indicate that PAF-AH passes through the acidic stomach environment into small intestine where it retains functional activity and avoids absorption into the systemic circulation. Because PAF-AH activity is deficient in newborns (36), and activity is present in breast milk but not formula, enteral supplementation of this compound in premature formula warrants additional investigation.

Experimental evidence suggests that the risk factors for NEC stimulate the production of PAF that results in the development of intestinal necrosis. Hypoxia and feeding stimulate systemic and local PAF production in various models, and bacterial colonization with activation of endotoxin significantly increases PAF synthesis (32,37). Elevated intestinal PAF concentrations in patients (newborns) unable to degrade the phospholipid adequately (because of a developmental deficiency) increases the theoretical risk for development of NEC. Many studies have shown that PAF stimulates the production of a variety of secondary mediators, including TNF, IL-1, NO, oxygen radicals, thromboxanes, and leukotrienes, and that these compounds in turn are associated with the propagation of intestinal necrosis.

Although the evidence strongly supports a key role for PAF in the initiation of NEC, the biochemical mechanisms for these events are poorly delineated. We hypothesize that PAF and other mediators activate apoptosis (programmed cell death) of intestinal epithelial cells that results in altered tight junctional integrity and increased mucosal permeability. As a result, excessive bacterial translocation follows with the activation of the secondary inflammatory cascade, and ultimately, NEC. Assessment of the mechanisms that regulate PAF-induced intestinal necrosis are under current investigation.


Use of anaerobic bacterial supplementation in the treatment or prevention of gastrointestinal disease has been well described. Studies with bifidobacteria or lactobacilli colonization have shown a reduction in diarrheal disease associated with rotavirus in children and with international travel in adults (38–40). Nonetheless, the effects of these bacteria in neonates have not been well studied. Bifidobacteria species are the most predominant organisms in the intestinal tract of healthy breast-fed infants, whereas premature neonatal intensive care patients have a relative absence of these anaerobic bacteria (41).

Therefore, to investigate the role of bifidobacteria supplementation on neonatal NEC, we used the rat model described earlier. Neonatal Sprague–Dawley rats are delivered through abdominal incision at 21.5 days to avoid exposure to breast milk feedings. Animals are then fed artificial formula (200 kcal/kg per day) through an orogastric feeding catheter every 3 hours and exposed to asphyxia twice daily (50 seconds of nitrogen exposure followed by 10 minutes in 4°C). Following this protocol, 70% to 80% of animals show development of clinical and pathologic characteristics of NEC by 96 hours of life. To study the effects of bifidobacteria supplementation, we treated animals with 109Bifidobacterium infantis (Bifido) organisms per day through the orogastric feeding tube, and measured the degree of colonization, activation of the inflammatory cascade, and the incidence of NEC. We found that Bifido supplementation resulted in colonization of stool and intestinal lumen by 24 and 48 hours, whereas placebo-treated animals had no detectable Bifido colonization. Furthermore, Bifido colonization reduced the incidence of NEC and death compared with placebo-treated animals (NEC: 5/22 versus 18/26 placebo;P < 0.05; see Table 1). In addition, we found that Bifido treatment reduced endotoxinemia (mean value: 190 endotoxin unit (EU) control versus 21 EU Bifido;P < 0.01) and intestinal PLA2 gene expression compared with control animals. Therefore, it appears that bifidobacteria colonization results in less risk for NEC through the modulation of the inflammatory cascade.

Effect of formula supplementation with Bifidobacterium infantis on disease outcomes in a neonatal model of necrotizing enterocolitis

The specific mechanisms responsible for bifidobacteria protection in NEC are still undetermined. As shown in Figure 1, probiotics can alter several key components of intestinal inflammation, including gut bacterial colonization, production of inflammatory mediator compounds, activation of the intestinal immune system, and the integrity of the mucosal barrier (2). The specific beneficial role of these organisms has been attributed to modulation of microflora growth and adherence, production of substances toxic to aerobic bacteria, reduction of intraluminal pH, modulation of the immune response, promotion of mucosal barrier function, and reduction of mucosal permeability. Gram-negative bacteria colonization after probiotic supplementation has been shown to be decreased in some studies but no different in others (4,42–44). Investigators have suggested that this difference could be attributable to the dose regimen of bacteria, or timing and duration of supplementation. In one report, Bifido supplementation inhibited bacterial translocation in mice in baseline conditions but had no effect after serious thermal injury (3). Specific studies investigating the effects of probiotic supplementation have shown increased fecal IgA in children (45) and adults (46), increased T-cell and macrophage production of cytokines (47), and increased blood leukocyte phagocytosis in humans (48). In a study of Clostridium- induced experimental NEC in quails, Bifido supplementation reduced the incidence of intestinal disease from 8 of 11 control subjects to 0 of 8 with probiotic supplementation (5). These researchers suggested that butyric acid production is a critical factor regulating the pathophysiologic effects. In summary, several mechanisms may be responsible for the protective effects of probiotics in intestinal inflammatory diseases and experimental NEC (Fig. 2).

FIG. 1.
FIG. 1.:
Proposed roles of probiotics in the functions of small intestine. Probiotics have been proposed to regulate intestinal bacterial colonization in favor of nonpathogenic organisms, to release chemical mediators improving the differentiation and barrier function of gut epithelium and the development of mucosal immunity.
FIG. 2.
FIG. 2.:
Schematic of the proposed mechanisms that may underlie the pathophysiologic course of necrotizing enterocolitis. Intermittent ischemic episodes, and the absence of protective factors from formula lead to a mucosal injury that is worsened by the colonization of pathologic bacteria and by the uncontrolled release of inflammatory mediators by the premature mucosal immune system. Host defense mechanisms against the bacteria provide some protection but may also contribute to the mucosal injury by the release of inflammatory mediators. Bifidobacteria may hinder the colonization of pathogenic bacteria and improve the differentiation and development of mucosal immunity.

There is limited experience with probiotic supplementation in human newborns. Bifidobacterium bifidum– supplemented formula was compared with breast milk and formula-feeding in healthy full-term newborns (49). In this report, supplementation resulted in reduced fecal pH and a degree of bifidobacteria colonization at 1 month similar to that with human milk feeding but different from that with formula-feeding alone. Recently, Bifidobacterium breve supplementation has been studied in premature infants (50). The study involved a randomized, controlled trial of 109 organisms per day with clinical follow-up and quantitative microbiologic analyses. The investigators found that infants consuming supplemented feedings had higher rates of fecal bifidobacteria colonization (73% versus 12%), decreased gastric aspirates, and improved weight gain and feeding tolerance with no identified side effects. Although 91 premature infants were studied, the incidence or severity of NEC was not reported. In summary, although probiotic supplementation appears to be well tolerated in full-term and premature newborns, determining the effects on intestinal health and NEC requires larger, well-controlled, clinical trials.


In conclusion, NEC remains an important cause of morbidity and mortality in premature infants throughout the world. Although the pathophysiologic course is not yet fully elucidated, it appears that the biochemical mediator, PAF plays a key role in the inflammatory necrosis observed in histologic sections from humans and experimental animals. Many mechanisms may account for the beneficial effects observed after bifidobacteria supplementation in models of intestinal inflammation in humans and animals. Enteral probiotic supplementation to premature newborns with patterns of intestinal colonization different from those in healthy, full-term newborns may modulate the process of intestinal inflammation and necrosis and reduce the incidence of NEC in this population.


1. Furlano RI, Walker WA. Immaturity of gastrointestinal host defense in newborns and gastrointestinal disease states. Adv Pediatr 1998; 45:201–22.
2. Walker WA, Duffy LC. Diet and bacterial colonization: Role of probiotics and prebiotics. J Nutr Biochem 1998; 9:668–75.
3. Suzuki T, Itoh K, Kaneko T, Suzuki H. Inhibition of bacterial translocation from the gastrointestinal tract of mice by oral administration of a culture condensate of Bifidobacterium longum. J Vet Med Sci 1997; 59:665–9.
4. Gibson GR, Wang X. Regulatory effects of bifidobacteria on the growth of other colonic bacteria. J Appl Bacteriol 1994; 77:412–20.
5. Butel MJ, Roland N, Hibert A, et al. Clostridial pathogenicity in experimental necrotising enterocolitis in gnotobiotic quails and protective role of bifidobacteria. J Med Microbiol 1998; 47:391–9.
6. Adawi D, Kasravi FB, Molin G, Jeppsson B. Effect of Lactobacillus supplementation with and without arginine on liver damage and bacterial translocation in an acute liver injury model in the rat. Hepatology 1997; 25:642–7.
7. Kliegman RM, Fanaroff AA. Necrotizing enterocolitis (review). N Engl J Med 1984; 310:1093–103.
8. Walsh MC, Kliegman RM. Necrotizing enterocolitis: Treatment based on staging criteria. Pediatr Clin North Am 1986; 33:179–201.
9. Uauy RD, Fanaroff AA, Korones SB, Phillips EA, Phillips JB, Wright LL. Necrotizing enterocolitis in very low birth weight infants: Biodemographic and clinical correlates. National Institute of Child Health and Human Development Neonatal Research Network. J Pediatr 1991; 119:630–8.
10. Siu YK, Ng PC, Fung SC, et al. Double blind, randomised, placebo controlled study of oral vancomycin in prevention of necrotising enterocolitis in preterm, very low birthweight infants. Arch Dis Child Fetal Neonatal Ed 1998; 79:F105–9.
11. Halac E, Halac J, Begue EF, et al. Prenatal and postnatal corticosteroid therapy to prevent neonatal necrotizing enterocolitis: A controlled trial. J Pediatr 1990; 117:132–8.
12. Eibl MM, Wolf HM, Furnkranz H, Rosenkranz A. Prevention of necrotizing enterocolitis in low-birth-weight infants by IgA-IgG feeding. N Engl J Med 1988; 319:1–7.
13. Ichihashi H, Nagasawa H, Kuwabara N, et al. Early enteral feeding of neonates less than 1000 gram birth weight. J Jpn Neonat Assoc 1998; 34:589–94.
14. Walsh MC, Kliegman RM, Hack M. Severity of necrotizing enterocolitis: Influence on outcome at 2 years of age. Pediatrics 1989; 84:808–14.
15. Caplan MS, MacKendrick W. Inflammatory mediators and intestinal injury (review). Clin Perinatol 1994; 21:235–46.
16. Udall JN Jr. Gastrointestinal host defense and necrotizing enterocolitis. J Pediatr 1990; 117:S33–43.
17. Nowicki PT, Hansen NB, Hayes JR, Menke JA, Miller RR. Intestinal blood flow and O2 uptake during hypoxemia in the newborn piglet. Am J Physiol 1986; 251:G19–24.
18. Nowicki PT, Miller CE. Autoregulation in the developing postnatal intestinal circulation. Am J Physiol 1988; 254:G189–93.
19. Stoll BJ, Kanto WP Jr Glass RI, Nahmias AJ, Brann AW Jr. Epidemiology of necrotizing enterocolitis: A case control study. J Pediatr 1980; 96:447–51.
20. Grosfeld JL, Chaet M, Molinari F, et al. Increased risk of necrotizing enterocolitis in premature infants with patent ductus arteriosus treated with indomethacin. Ann Surg 1996; 224:350–7.
21. Lucas A, Cole TJ. Breast milk and neonatal necrotising enterocolitis [see comments]. Lancet 1990; 336:1519–23.
22. Brown EG, Sweet AY. Preventing necrotizing enterocolitis in neonates. JAMA 1978; 240:2452–4.
23. Lawrence G, Bates J, Gaul A. Pathogenesis of neonatal necrotising enterocolitis. Lancet 1982; 1:137–9.
24. Gonzalez–Crussi F, Hsueh W. Experimental model of ischemic bowel necrosis: The role of platelet-activating factor and endotoxin. Am J Pathol 1983; 112:127–35.
25. Sun XM, Hsueh W. Bowel necrosis induced by tumor necrosis factor in rats is mediated by platelet-activating factor. J Clin Invest 1988; 81:1328–31.
26. Hsueh W, Gonzalez–Crussi F, Arroyave JL, Anderson RC, Lee ML, Houlihan WJ. Platelet activating factor-induced ischemic bowel necrosis: The effect of PAF antagonists. Eur J Pharmacol 1986; 123:79–83.
27. Caplan MS, Kelly A, Hsueh W. Endotoxin and hypoxia-induced intestinal necrosis in rats: The role of platelet activating factor. Pediatr Res 1992; 31:428–34.
28. Caplan MS, Sun XM, Hseuh W, Hageman JR. Role of platelet activating factor and tumor necrosis factor-alpha in neonatal necrotizing enterocolitis. J Pediatr 1990; 116:960–4.
29. Farr RS, Wardlow ML, Cox CP, Meng KE, Greene DE. Human serum acid-labile factor is an acylhydrolase that inactivates platelet-activating factor. Fed Proc 1983; 42:3120–2.
30. Mukherjee AB, Miele L, Pattabiraman N. Phospholipase A2 enzymes: Regulation and physiological role (review). Biochem Pharmacol 1994; 48:1–10.
31. Wang H, Tan X, Chang H, Gonzalez–Crussi F, Remick DG, Hsueh W. Regulation of platelet-activating factor receptor gene expression in vivo by endotoxin, platelet-activating factor and endogenous tumour necrosis factor. Biochem J 1997; 322:603–8.
32. Caplan MS, Adler L, Kelly A, Hsueh W. Hypoxia increases stimulus-induced PAF production and release from human umbilical vein endothelial cells. Biochim Biophys Acta 1992; 1128:205–10.
33. Caplan MS, Hedlund E, Adler L, Hsueh W. Role of asphyxia and feeding in a neonatal rat model of necrotizing enterocolitis. Pediatr Pathol 1994; 14:1017–28.
34. Caplan MS, Hedlund E, Adler L, Lickerman M, Hsueh W. The platelet-activating factor receptor antagonist WEB 2170 prevents neonatal necrotizing enterocolitis in rats. J Pediatr Gastroenterol Nutr 1997; 24:296–301.
35. Caplan MS, Lickerman M, Adler L, Dietsch GN, Yu A. The role of recombinant platelet-activating factor acetylhydrolase in a neonatal rat model of necrotizing enterocolitis. Pediatr Res 1997; 42:779–83.
36. Caplan M, Hsueh W, Kelly A, Donovan M. Serum PAF acetylhydrolase increases during neonatal maturation. Prostaglandins 1990; 39:705–14.
37. MacKendrick W, Hill N, Hsueh W, Caplan M. Increase in plasma platelet-activating factor levels in enterally fed preterm infants. Biol Neonate 1993; 64:89–95.
38. Saavedra JM, Bauman NA, Oung I, Perman JA, Yolken RH. Feeding of Bifidobacterium bifidum and Streptococcus thermophilus to infants in hospital for prevention of diarrhoea and shedding of rotavirus. Lancet 1994; 344:1046–9.
39. Oksanen PJ, Salminen S, Saxelin M, et al. Prevention of travellers' diarrhoea by Lactobacillus GG. Ann Med 1990; 22:53–6.
40. Kaila M, Isolauri E, Saxelin M, Arvilommi H, Vesikari T. Viable versus inactivated lactobacillus strain GG in acute rotavirus diarrhoea. Arch Dis Child 1995; 72:51–3.
41. Bennet R, Nord CE, Zetterstrom R. Transient colonization of the gut of newborn infants by orally administered bifidobacteria and lactobacilli. Acta Paediatr 1992; 81:784–7.
42. Kinouchi T, Kataoka K, Bing SR, et al. Culture supernatants of Lactobacillus acidophilus and Bifidobacterium adolescentis repress ileal ulcer formation in rats treated with a nonsteroidal antiinflammatory drug by suppressing unbalanced growth of aerobic bacteria and lipid peroxidation. Microbiol Immunol 1998; 42:347–55.
43. Romond MB, Haddou Z, Mialcareck C, Romond C. Bifidobacteria and human health: Regulatory effect of indigenous bifidobacteria on Escherichia coli intestinal colonization. Anaerobe 1997; 3:131–6.
44. Abe F, Momosa H, Igarashi M, et al. The effect of administration of bifidobacteria on the intestinal flora and growth of newborn piglets. J Gen Appl Microbiol 1996; 42:257–62.
45. Fukushima Y, Kawata Y, Hara H, Terada A, Mitsuoka T. Effect of a probiotic formula on intestinal immunoglobulin A production in healthy children. Int J Food Microbiol 1998; 42:39–44.
46. Link–Amster H, Rochat F, Saudan KY, Mignot O, Aeschlimann JM. Modulation of a specific humoral immune response and changes in intestinal flora mediated through fermented milk intake (published corrections appear in FEMS Immunol Med Microbiol 1995;12:83 and 1995;12:273). FEMS Immunol Med Microbiol 1994; 10:55–63.
47. Marin ML, Tejada–Simon MV, Lee JH, Murtha J, Ustunol Z, Pestka JJ. Stimulation of cytokine production in clonal macrophage and T-cell models by Streptococcus thermophilus: Comparison with Bifidobacterium sp. and Lactobacillus bulgaricus. J Food Prot 1998; 61:859–64.
48. Schiffrin EJ, Rochat F, Link–Amster H, Aeschlimann JM, Donnet–Hughes A. Immunomodulation of human blood cells following the ingestion of lactic acid bacteria. J Dairy Sci 1995; 78:491–7.
49. Langhendries JP, Detry J, Van Hees J, et al. Effect of a fermented infant formula containing viable bifidobacteria on the fecal flora composition and pH of healthy full-term infants (see comments). J Pediatr Gastroenterol Nutr 1995; 21:177–81.
50. Kitajima H, Sumida Y, Tanaka R, Yuki N, Takayama H, Fujimura M. Early administration of Bifidobacterium breve to preterm infants: Randomised controlled trial. Arch Dis Child Fetal Neonatal Ed 1997; 76:F101–7.
© 2000 Lippincott Williams & Wilkins, Inc.