Therapeutic Use of Prebiotics, Probiotics, and Postbiotics to Prevent Necrotizing Enterocolitis: What is the Current Evidence?

Synopsis

Necrotizing enterocolitis (NEC) is a leading cause of neonatal morbidity and mortality and preventative therapies that are both effective and safe are urgently needed. Current evidence from therapeutic trials suggests that probiotics are effective in decreasing NEC in preterm infants and probiotics are currently the most promising therapy on the horizon for this devastating disease. However, concerns regarding safety and optimal dosing have limited the widespread adoption of routine clinical use of probiotics in preterm infants. In addition, prebiotics and postbiotics may be potential alternatives or adjunctive therapies to the administration of live microorganisms, although studies demonstrating their clinical efficacy in preventing NEC are lacking. This review summarizes the current evidence regarding the use of probiotics, prebiotics and postbiotics in the preterm infant, including its therapeutic role in preventing NEC.

Introduction: Necrotizing Enterocolitis

Necrotizing enterocolitis (NEC) is a leading cause of neonatal morbidity and mortality and the most common gastrointestinal emergency in neonates [1, 2]. NEC develops in approximately 1 out of 10 infants born at less than 29 weeks gestation [3]. Approximately 20% to 30% of VLBW (birth weight ≤ 1500 grams) infants who develop NEC die [4] and those infants who survive the disease are at risk for long-term complications, including neurodevelopmental impairment, short bowel syndrome, and impaired growth [5]. This results in considerable health costs, with the financial impact of affected infants in the United States estimated to be between $500 million to $1 billion per year [6]. Finally, NEC is particularly poignant for families because it affects preterm infants who have survived the initial postnatal period only to develop a new and life-threatening illness.

NEC is characterized by intestinal injury, inflammation and necrosis. Despite investigative efforts, the underlying pathogenesis remains unclear. Leading hypotheses implicate a multifactorial pathophysiology, which includes host factors, inflammatory propensity of the immature gut, enteral feeding, and abnormal bacterial colonization [6–8] (Box 1). NEC can rapidly progress from early clinical signs to extensive intestinal necrosis within hours, limiting the effectiveness of therapeutic intervention. As such, approaches to prevent NEC have become a focus of research efforts and preventative therapies for NEC are urgently needed [9].

Box 1. Risk Factors Influencing NEC Predisposition.

• Prematurity

  • Inflammatory propensity of the immature gut [74–76]

  • Decreased intestinal barrier function [77, 78]

  • Decreased gut motility and aberrant vascular regulation [79, 80]

• Enteral feeding

  • Aggressive advancement of feeding [81, 82]

  • Non-human milk feeding [83, 84]

• Abnormal bacterial colonization

  • Prolonged empiric initial antibiotic therapy [21, 23, 24]

  • Decreased commensal flora [20, 85]

  • Increased pathogenic bacteria [19, 85]

Probiotics are the most promising treatment on the horizon for this devastating disease. However, as we review in this article, further study is needed before probiotics can be routinely recommended as a preventative therapy. In addition, prebiotics and postbiotics remain potential alternatives or adjunctive therapies to the use of live microorganisms. Herein, we review the role of microorganisms in intestinal heath and disease in neonates and the current evidence supporting prebiotic, probiotic and postbiotic use in preterm infants. In addition, we discuss the potential mechanisms of action of probiotic organisms and weigh the risks and benefits of therapy. Important terminology is defined in Box 2.

Box 2. Key Definitions.

• Probiotics are supplements or foods that contain viable microorganisms that alter the microflora of the host.

• Prebiotics are supplements or foods that contain a nondigestible ingredient that selectively stimulates the growth and/or activity of indigenous bacteria.

• Postbiotics are non-viable bacterial products or metabolic byproducts from probiotic microorganisms that have biologic activity in the host.

• Synbiotic is a product that contains both probiotics and prebiotics.

Role of intestinal microorganisms in intestinal development and health

The neonatal intestine is the largest interface of the host to the external environment. Ideally, the intestine protects the vulnerable neonate from pathogens and toxins while allowing harmonious habitation of commensal bacteria. In addition, the intestine must protect itself from bacterial pathogens that cause inflammation while encouraging the growth and habitation of commensal bacteria, which can attenuate inflammation and maintain intestinal homeostasis [10]. Importantly, uncontrolled inflammation can lead to tissue injury and necrosis and is thought to play an important role in the pathogenesis of NEC [6].

The immature neonatal gut in the preterm infant has even greater challenges as it transitions from a sterile lumen absent of microbes to the fully realized “bioreactor” of the mature adult gut. Very soon after birth, a newborn infant’s intestine begins to acquire a diverse set of commensal bacteria. These bacteria eventually grow to outnumber their host by a factor of ten to one [11–13]. The infant/host provides an hospitable, temperature-stable, nutrient-rich environment for bacteria while receiving, in return, benefits from the commensal bacteria, which include digestion, absorption, and storage of nutrients [14, 15], intestinal homeostasis and protection against injury [13, 16], and the development and control of epithelial immune responses and function [17, 18]. However, this postnatal transition from a sterile to commensal-rich intestine may become disrupted or delayed in infants born prematurely.

Abnormal bacterial colonization in preterm infants

Premature infants, when compared to their term counterparts, are more likely to be colonized with fewer, more virulent organisms [19]. In addition, they have delayed acquisition of commensal bacteria, particularly Bifidobacteria [20]. Why does this occur? Several explanations are likely (Figure 1). Preterm infants and their mothers are commonly exposed to antibiotic therapy, both of which have been associated with an increased risk of NEC [21–24]. In addition, since preterm infants are often delivered by cesarean and experience delayed enteral feeding, they are less likely to acquire commensal flora perinatally from passage through the birth canal or from human milk feeding. This may lead to decreased colonization of beneficial probiotic bacteria, including Bifidobacteria, Lactobacillus and Bacteroides [25, 26]. The hospital environment, with its preponderance of pathogenic organisms [27], may also negatively affect the intestinal colonization of beneficial commensal bacteria. Ongoing studies investigating the intestinal microbiome in premature neonates will provide greater insight into the role of microbial diversity on the development of NEC [28–30]. Since abnormal bacterial colonization likely plays a role in the pathogenesis of NEC, probiotic bacteria may exert their beneficial effects by restoring or supplying the essential commensal strains necessary for protection against intestinal inflammation and injury. Prebiotics and postbiotics may also accomplish this goal, through the promotion of commensal growth or by mirroring commensal activity, respectively.

Figure 1.

Factors influencing abnormal intestinal bacterial colonization in preterm infants.

Mechanisms of Action

Probiotics

What is the mechanism of action for probiotics? In preterm infants, probiotic supplementation can allow acquisition of normal commensal flora in a host where this process has been delayed or support the transition to an intestinal microbiome with beneficial microbes, particularly in hosts where this process has been disrupted. Several mechanisms of probiotic action may explain how their therapeutic use can help prevent NEC. These mechanisms include enhancement of epithelial barrier function, competitive exclusion of pathogens, and direct anti-inflammatory effects on epithelial signaling pathways [31–33]. At the cellular level, probiotics have a number of important effects (Figure 2): 1) attenuation of NF-κB activation, a major pro-inflammatory pathway [34]; 2) upregulation of cytoprotective genes [13, 35]; 3) prevention of apoptosis and cell death [35, 36]; 4) generation of reactive oxygen species important in cell signaling [37, 38]; 5) Induction of the expression of tight junction proteins necessary for barrier function [39, 40]. Whether live microorganisms, instead of killed or inactivated bacteria or bacterial products, are required for these beneficial effects remains an important area of study and recent data suggest that bacterial products, in the absence of viable organisms, may have similar effects on signaling pathways [41] and barrier function [39]. These bacterial products, which we broadly characterize as postbiotics, are discussed in a separate section below.

Figure 2.

Mechanisms of action of probiotics at the cellular level in intestinal epithelia.

Since the preterm gut demonstrates delayed commensal colonization and low bacterial diversity, it may be particularly amenable to therapeutic manipulation by probiotic administration. In keeping with this idea, a number of clinical studies have demonstrated the benefit of probiotic administration in reducing the incidence and severity of NEC (Table 1). Most of these trials have used strains of probiotics from the Genus Lactobacillus or Bifidobacteria, although the treatment regimen, including dose and duration of therapy vary widely (Table 2). In addition, concerns regarding the risks of therapy, optimal species or combination of species, duration of therapy and dosing remain unanswered. The therapeutic use of probiotics is discussed below.

Table 1.

Summary of Probiotic Therapy Trials in Preterm Infants

Study Authors	Year Published	Number of Infants	BW and/or GA	Primary Outcome of NEC
Kitajima et al.[86]	1997	91	<1500g	No
Dani et al.[87]	2002	585	<1500g or <33wk	Yes
Costalos et al.[88]	2003	87	28–32wk	No
Bin-Nun et al.[89]	2005	145	<1500g	Yes
Lin et al.[90]	2005	367	<1500g	Yes
Manzoni et al.[91]	2006	80	<1500g	No
Mohan et al.[92]	2006	38 of 69*	<37wk	No
Stratiki et al.[93]	2007	41 of 75*	27–36wk	No
Lin et al.[69]	2008	434	<1500g & <34wk	Yes
Samanta et al.[68]	2009	186	<1500g & <32wk	Yes
Rouge et al.[94]	2009	94	<1500g & <32wk	No
Manzoni et al.[67]	2009	319 of 472	<1500g	No
Underwood et al.[48]	2009	90	750–2000g & <35wk	No
Mihatsch et al.[95]	2010	183	<1500g and <30wk	No
Braga et al.[96]	2011	231	750–1499g	Yes
Sari et al.[97]	2011	221	<1500g or <33wk	Yes
Fernandez-Carrocera et al.[98]	2012	150	<1500g	Yes

Abbreviations: NEC, necrotizing enterocolitis; BW, birth weight; GA, gestational age; Data from individual studies referenced and from Deshpande et al.[66].

^*Only infants <34wk and <1500g included.

Table 2.

Summary of Probiotic Strains Used in Therapeutic Trials in Preterm Infants

Probiotic Genus	Species	Dose (CFU/day)	Reference
Bifidobacterium	bifidus	1 × 109 a	[48], [89]
		2 × 109 a	[69]
		5 × 109 a	[68]
	breve	0.5 × 109	[86]
		3.5 × 107–3.5 × 109 a	[96]
	infantis	2.76 × 107 a	[98]
		1 × 109 a	[48], [89]
		5 × 109 a	[68]
		NR	[90]
	lactis	2 × 107 b	[93]
		1.6–4.8 × 109	[92]
		1.2 × 1010	[95]
	longus	4 × 108 a	[94]
		1 × 109 a	[48]
		5 × 109 a	[68]
Lactobacillus	acidophilus	1 × 109 a	[48], [98]
		2 × 109 a	[69]
		5 × 109 a	[68]
		NR	[90]
	casei	1 × 109 a	[98]
		3.5 × 107–3.5 × 109 a	[96]
	rhamnosus GG	4 × 108	[94]
		4.4 × 108 a	[98]
		1 × 109	[48]
		6 × 109	[67], [87], [91]
	plantarum	1.76 × 108 a	[98]
	sporogenes	3.5 × 108	[97]
Streptococcus	theromphilus	6.6 × 105 a	[98]
		1 × 109 a	[89]
Saccharomyces	boulardii	2 × 109 per kg	[88]

Abbreviations: CFU, colony forming units; NR, not reported in CFU per day

^aIndicates therapeutic use in combination with other probiotic species

^bDose is per gram of milk powder of preterm formula

Prebiotics

Another potential therapeutic strategy to prevent NEC that may carry less infectious risk is prebiotic therapy. Prebiotics are non-digestible dietary products, which improve intestinal health by promoting selective growth of beneficial commensal bacteria [42]. The most common prebiotics are oligosaccharides, which are found in human milk. Prebiotics have been added to infant formulas in Japan for over two decades and there is increasing use of prebiotics in the commercial formula market in the United States although concerns regarding their efficacy remain [43, 44]. Several studies have demonstrated increased Bifidobacteria and decreased pathogenic bacteria in the stool of preterm infants fed prebiotic-containing formula with short-chain galacto-oligosaccharides (GOS) and/or fructo-oligosaccharides (FOS) as compared to control infants [45–47]. Similar effects of increased Bifidobacteria colonization were seen with the use of synbiotics, which combine prebiotics with probiotics [48]. In addition to promoting the growth of beneficial commensals, prebiotic therapy may also improve intestinal motility and gastric emptying, resulting in improved feeding tolerance [49, 50]. Evidence suggests that this effect of prebiotics on gastrointestinal motility is mediated by bacterial metabolites such as short-chain fatty acids[51], and, therefore, may actually be a postbiotic effect rather than a direct effect of the prebiotic. Term infants exhibit decreased stool pH and increased stool short-chain fatty acids when fed prebiotic containing formula, reportedly comparable to that seen in breastfed infants [52]. In addition, preliminary evidence suggests a positive effect on overall immune function [53]. While no serious adverse effects of prebiotic therapy were noted in the previously mentioned studies, prebiotic therapy is associated with several gastrointestinal side effects. These include flatulence, bloating, and diarrhea, which can be reversed by stopping treatment [42].

Postbiotics

Using bacterial products or metabolites remains an exciting, yet largely unexplored, therapeutic option for preventing NEC. Postbiotics aim to mimic the beneficial therapeutic effects of probiotics while avoiding the risk of administering live microorganisms to preterm infants with immature intestinal barriers or impaired immune defenses. Many commensal bacteria produce butyrate, a short-chain fatty acid produced by the catabolism of undigested carbohydrates in the intestine. Butryate is a major energy source for the colon and has an important role in intestinal growth and differentiation [54, 55], inflammatory suppression [56–58] (including suppression through generation of reactive oxygen species)[59], and regulation of apoptosis [60, 61]. It has been used with limited success in inflammatory bowel disease [62]. Perhaps this and other small molecule products of the normal flora are at least partially responsible for the beneficial effects of the normal flora (and exogenous pro- and prebiotics), and could be employed as a more controllable and safer therapeutic surrogate. Heat-killed probiotics may also function, in the broad sense, as postbiotics. Heat-killed microorganisms retain important bacterial structures that may exert biologic activity in the host. We recently reported that heat-killed LGG has similar beneficial efficacy as live probiotics in accelerating intestinal barrier maturation in a murine model of immature intestines [39]. More importantly, while administration of higher doses of live LGG led to increased mortality in immature neonatal mice, administration of heat-killed LGG did not.

Therapeutic Options

Numerous probiotic organisms have been studied in preterm infants, at varied dosages and durations of therapy (Table 2). The diversity of use has made selecting an optimal probiotic regimen difficult. Several studies have used single agents, while a number of larger trials have used a combination of probiotics. Dosing varies among probiotic species and ranges from 10⁵ to 10¹⁰ CFU per day. In addition, the actual probiotic composition may vary from what is advertised [48]. Most studies randomized infants and/or initiated therapy within a week after birth and duration of therapy typically lasted beyond 1 month of age with several studies continuing therapy until hospital discharge. The two most commonly used probiotic strains were from the genera Bifidobacteria and Lactobacillus. A meta-analysis by Wang et al. [63], including a number of recent trials conducted in China, compared the relative efficacy of these strains by pooling studies that used these probiotic agents and found that Bifidobacteria, Lactobacillus and a combination of the two strains had similar efficacy in the prevention of NEC (Table 3). Of note, monitoring for adverse effects varied greatly among the studies, particularly in those where NEC was not a primary outcome measure.

Table 3.

Comparison of the Efficacy of Probiotic Strains

Outcome	Probiotic	Total Patients (n)	Pooled Relative Risk (95% CI)
NEC	Lactobacillus	1205	0.37 (0.19–0.73)
	Bifidobacteria	976	0.30 (0.16–0.58)
	Both	1403	0.33 (0.19–0.58)
Death	Lactobacillus	1205	0.61 (0.38–0.97)
	Bifidobacteria	340	0.74 (0.18–2.97)
	Both	1312	0.47 (0.26–0.87)
Sepsis	Lactobacillus	1205	0.79 (0.46–1.36)
	Bifidobacteria	340	0.84 (0.29–2.41)
	Both	1312	0.90 (0.60–1.36)

Abbreviations: NEC, necrotizing enterocolitis; CI, confidence interval Adapted from Wang et al.[63]

For prebiotics, several oligosaccharides including fructose, galactose, lactulose, inulin, or a combinations of these are potential therapeutic options for clinical use [64]. GOS, FOS and a combination of the two (GOS-FOS) have been used clinically in preterm infants and studies have shown increased Bifidobacteria stool colony counts in prebiotic treated infants, although none of these studies evaluated NEC as a primary outcome measure [65].

Clinical Outcomes

Clinical trials have consistently demonstrated trends in favor of probiotic treatment, compared to placebo, in reducing the incidence of NEC (Figure 3A). Four of these studies demonstrated a statistically significant benefit and two recent meta-analyses of therapeutic trials have both shown similar pooled relative risks in favor of probiotics for the prevention of NEC. Deshpande et al. evaluated 11 trials and found an overall pooled relative risk (RR) of 0.35 (95% CI 0.23–0.55) for NEC in infants randomized to probiotic therapy compared to controls [66]. Wang et al., as mentioned previously, included more recent trials, and found a similar pooled RR of 0.33 (95% CI 0.24–0.46) from a total of 20 studies [63]. Of note, only 8 of 17 trials we identified had NEC listed as a primary outcome measure and the dose and species of probiotics used varied widely (Tables 1 and 2). Importantly, the previously mentioned meta-analyses of also demonstrated a decreased risk of death (Figure 3B) in infants randomized to probiotic treatment compared to control. It is likely that the mortality benefit is through a reduction in NEC- and sepsis-related deaths, although the causes of death have not been routinely evaluated in clinical trials.

Figure 3.

Plots demonstrate the relative risk of necrotizing enterocolitis (NEC) (panel A) and mortality (panel B) in the probiotic treatment arm, compared to placebo, of therapeutic trials in preterm infants. Error bars reflect 95% confidence interval (CI) for relative risk estimates. Two studies had no NEC events in the probiotic treatment arm and estimates, without 95% CI, are indicated by black arrows. Data from individual studies referenced as well as from Deshpande et al.[66] and Wang et al.[63].

However, the benefit for probiotics in the prevention of sepsis (Figure 4) is not as clear. Two randomized trials, including a large trial by Manzoni et al., have shown a benefit of probiotics in reducing late-onset sepsis [67, 68]. However, Manzoni et al. used the probiotic LGG in combination with lactoferrin and did not evaluate probiotic treatment alone. Other studies have shown no benefit and one study showed an increased risk of sepsis with a RR 1.67 (95% CI 1.04–2.67), raising concerns regarding the risk of sepsis with probiotic therapy [69].

Figure 4.

Plot demonstrates the relative risk of sepsis in the probiotic treatment arm, compared to placebo, of therapeutic trials in preterm infants. Error bars reflect 95% confidence interval for relative risk estimates. Data from individual studies referenced as well as from Deshpande et al.[66] and Wang et al.[63].

While prebiotics and postbiotics are potential alternatives or supplements to probiotic therapy, no clinical trials have studied the therapeutic use of these substances for the prevention of NEC. Therefore, we can only speculate on the potential benefit of prebiotics and postbiotics in the preterm infant until more definitive data are available regarding their effects on important clinical outcomes of prematurity, such as NEC, sepsis or death.

Complications and Concerns

Although the clinical efficacy of probiotics in preventing NEC is promising, concerns regarding the complications associated with therapy have mitigated widespread use. Probiotic use in premature infants could expose intestinal epithelia with poor defenses and a tendency towards inflammation to a microbial challenge too soon, resulting in inflammation, injury or sepsis. Several reports of probiotic-associated sepsis [70–72] have raised concerns regarding routine clinical use of live bacteria in hosts, such as premature infants, who have immature epithelial barrier defenses. The American Academy of Pediatrics Committee on Nutrition - Section On Gastroenterology, Hepatology, And Nutrition [73] and the ESPGHAN Committee on Nutrition [44] both have highlighted the need for large, well-designed clinical research studies before widespread use is adopted. The use of prebiotics or postbiotics may decrease these concerns regarding safety. However, no data from trials of prebiotics or postbiotics for the prevention of NEC in preterm infants are currently available. In addition, these agents are likely to encounter similar issues regarding optimal dosing as seen with probiotic therapy.

Summary

NEC remains a devastating complication of prematurity and new preventative therapies are urgently needed. Probiotic therapy has demonstrated promising efficacy for the prevention of NEC, as evidenced by a very strong treatment effect in favor of probiotic therapy in two recent meta-analyses [63, 66]. However, safety and dosing concerns continue to temper widespread use and these concerns are likely to remain until large, multicenter trials adequately designed to address safety are completed. In addition, the use of prebiotics or postbiotics may be potential alternatives to the use of live probiotic organisms, although the therapeutic benefit for these products in preventing NEC remains largely unknown. Further studies are needed to guide clinicians on the use of probiotics, prebiotics and postbiotics for the prevention of NEC.

Key Points.

• Current evidence suggests that probiotics are effective in decreasing NEC in preterm infants.

• Concerns regarding safety and optimal dosing have limited the routine clinical use of probiotics in preterm infants.

• Prebiotics and postbiotics are potential alternatives or adjunctive therapies to the administration of live microorganisms, although studies demonstrating their clinical efficacy in preventing NEC are currently lacking.