Neonatal necrotizing enterocolitis: Prevention

INTRODUCTION — Necrotizing enterocolitis (NEC) is one of the most common gastrointestinal (GI) emergencies in the newborn infant. It is a disorder characterized by ischemic necrosis of the intestinal mucosa, which is associated with severe inflammation, invasion of enteric gas-forming organisms, and dissection of gas into the intestinal wall and portal venous system [1]. Although early recognition and aggressive treatment of this disorder has improved clinical outcomes, NEC accounts for substantial long-term morbidity in survivors of neonatal intensive care, particularly in preterm very low birth weight (VLBW) infants (BW <1500 g). As a result, preventive efforts have focused on identifying interventions that will reduce the risk and severity of this disorder.

The prevention of NEC will be reviewed here. The pathology, pathogenesis, clinical features, diagnosis, and management of this disorder are discussed separately. (See "Neonatal necrotizing enterocolitis: Pathology and pathogenesis" and "Neonatal necrotizing enterocolitis: Clinical features and diagnosis" and "Neonatal necrotizing enterocolitis: Management and prognosis".)

EFFECTIVE INTERVENTIONS — Efforts to minimize the frequency or severity of NEC are directed at reducing exposure to risk factors and finding interventions that will prevent the disorder [2-6]. Since NEC occurs almost exclusively in preterm infants, prevention of preterm birth would have an impact on NEC incidence. Unfortunately, this goal has not yet been met.

Antenatal corticosteroids — Antenatal corticosteroids reduce the risk of respiratory distress syndrome (RDS) and mortality in preterm infants. In addition, exposure to antenatal corticosteroids reduces the risk of NEC, intraventricular hemorrhage, and retinopathy of prematurity [7]. As a result, antenatal corticosteroids should be given to all women at risk for preterm delivery within seven days. The use and benefit of antenatal corticosteroids are discussed in detail separately. (See "Antenatal corticosteroid therapy for reduction of neonatal respiratory morbidity and mortality from preterm delivery".)

Human milk feeding

●Our approach – In our center, we preferentially use human milk feeding for all preterm neonates. Whenever possible, mother's milk should be used as it provides the optimal benefits of human milk. The additional processing (pasteurization, freezing) and storage needed for donor human milk reduces some of the protective benefits that are provided by mother's milk. For infants up to 34 weeks adjusted gestational age, if mother's milk is not available, we provide pasteurized donor milk as a bridge until there are sufficient amounts of mother's milk. This practice is consistent with guidance from the American Academy of Pediatrics (AAP) [8-10].

Resources for supporting breast milk feeding in preterm infants, including use of donor milk, are provided through the Center for Breastfeeding Medicine and the California Perinatal Quality Care Collaborative Nutrition Toolkit.

Additional details regarding use of donor milk in preterm infants are provided separately. (See "Human milk feeding and fortification of human milk for premature infants", section on 'Use of donor milk'.)

●Benefits of human milk – Human milk feeding lowers the risk of NEC compared with intact bovine protein-based formula (figure 1) [6,11-15]. Mechanisms by which human milk is thought to reduce the risk of NEC include [16]:

•Lowers gastric pH.

•Enhances intestinal motility.

•Positively impacts the gut microbiota through antimicrobial effects and prebiotic and probiotic components, thereby lowering the risk of microbial dysbiosis, an important factor in the pathogenesis of NEC [17]. (See "Neonatal necrotizing enterocolitis: Pathology and pathogenesis", section on 'Susceptibility: Immunologic and intestinal immaturity' and "Neonatal necrotizing enterocolitis: Pathology and pathogenesis", section on 'Microbial dysbiosis'.)

The protective effect of human milk appears to be dose dependent, with higher intake of human milk leading to proportionally higher protection against developing NEC [18].

Additional details on the role of milk feeding in the pathogenesis of NEC and the benefits of human milk feeding for preterm infants are provided separately. (See "Neonatal necrotizing enterocolitis: Pathology and pathogenesis", section on 'Milk feeding' and "Human milk feeding and fortification of human milk for premature infants", section on 'Benefits of mother's milk'.)

●Evidence supporting donor milk – In a meta-analysis of nine trials (1675 neonates), rates of NEC were lower in infants receiving donor breast milk compared with formula feeding (4 versus 7 percent; relative risk [RR] 0.53, 95% CI 0.35-0.81) [11]. Similar findings were reported in an earlier meta-analysis [19]. Cost analysis included in the earlier meta-analysis suggested that while overall costs were similar with both approaches, post-discharge costs were lower in the donor milk group [19].

Standardized feeding protocol — Standardized feeding protocols provide a consistent approach to initiation of minimal enteral (trophic) feedings, timing and rate of advancement of feeding, timing of fortification, addition of additives like vitamins and iron, and criteria when to withhold and restart feeds.

At our institution, we use several feeding protocols based upon birth weight (BW) that cover all preterm and term newborns. Oral colostrum care with mother's milk is given as soon as it is available after birth, with feeds starting within the first 24-48 hours. For extremely low birth weight infants (BW <1000g), trophic feeds are provided for the first three days before advancing feeds. For very low birth weight (VLBW) infants (BW <1500 g), the typical rate of advancement is 20 to 30 mL/kg/day. However, faster rates of advancement (up to 40 mL/kg/day) may be acceptable [20]. The approach to initiating and advancing enteral feeding in preterm neonates is discussed in greater detail separately. (See "Approach to enteral nutrition in the premature infant".)

Observational studies have shown that standardized feeding protocols are associated with lower rates of NEC [21,22]. In a meta-analysis of 15 observational studies (18,160 preterm neonates) comparing rates of NEC before and after implementation of standardized feeding protocols, rates were considerably lower after implementation (risk ratio 0.22, 95% CI 0.13-0.36) [21]. The protocols used in these studies varied substantially with regard to timing of initiation and timing and speed of advancement. This suggests that standardization rather than any specific feature of the protocol is the key aspect of this intervention. Although the underlying mechanism for the benefit of standardized feeding protocols is unknown, it may be that they reduce variability of feeding advancement amongst multiple providers and maintain a linear rate of feeding advancement rather than having periods of feeding acceleration or deceleration.

Feeding protocols need to consider:

●Timing and rate of advancement of feeding – Trophic feeding provides infants very small amounts of milk to help prime the intestinal tract. The optimal timing of initiation of trophic feeding and volume of feeds for reducing NEC remains uncertain. However, based on the available data, early initiation and judicious advancement of feeds do not appear to increase the risk of NEC.

•Timing of initial trophic feeds – There have been concerns that initiation of enteral feeds within the first week of life for very preterm (<32 weeks, VPT) and VLBW infants would increase the risk of NEC. However, available data support that trophic feeding does not increase the incidence of NEC compared to enteral fasting [23-25]. A subsequent systematic review of very low birth weight VLBW infants also found that available data were insufficient to determine whether early tropic feeds (started before 96 hours postnatal age) were associated with a reduction in NEC compared to tropic feeds started after this period [26]. (See "Approach to enteral nutrition in the premature infant", section on 'Trophic feeds'.)

•Timing of progressive enteral feeds – A meta-analysis of VLBW infants demonstrated that delaying the introduction of progressive enteral feeds (after four days of age) was not associated with a reduced risk of NEC, and that the delay was associated with a longer time to establish full enteral feeds [27].

•Rate of advancement of enteral feeds – A systematic review of 14 trials of VLBW infants reported that faster rates of advancing feeds (30 to 40 mL/kg/day) compared with lower rates (15 to 24 mL/kg/day) did not increase the risk for NEC or mortality [20]. In a meta-analysis of 4206 neonates, similar rates of NEC were seen with faster rates compared with lower rates of advancement (0.054 versus 0.057 percent, relative risk [RR] 1.06, 95% CI 0.83-1.37). Similar results were also seen for mortality between the two advancement rates (0.07 versus 0.08, RR 1.13, 95% 0.91-1.39). A subsequent trial not included in the above meta-analysis also reported similar rates of NEC for VLBW infants whose feeds were advanced at faster versus slower rates [28]. (See "Approach to enteral nutrition in the premature infant", section on 'Rate of volume increase'.)

●Colostrum – Oral colostrum care (OCC) consists of administering drops of maternal colostrum into an infant's mouth within the first day of life. Most of these early volumes are absorbed directly in the mouth with some traveling to the gastrointestinal (GI) tract [29]. One meta-analysis reported that OCC may reduce the incidence of NEC, late onset sepsis, IVH and the time to full feeding in infants ≤32 weeks’ gestation [30]. However, this and previous meta-analyses are limited by the small number of patients and low-rate of NEC events [31,32]. Further research efforts are needed to determine whether there is any benefit of human colostrum in preventing NEC.

Other strategies — Other modifiable risk factors for NEC include prolonged exposure to systemic antibiotics, exposure to gastric acid suppressor therapy (eg, histamine 2 [H2] receptor blockers), and severe anemia/exposure to transfusion. While direct evidence is lacking to demonstrate that decreasing these risk factors lowers the incidence of NEC, avoiding unwarranted use of prolonged antibiotic therapy and/or H2 blockers are reasonable interventions to routinely implement in the neonatal intensive care unit. In addition, the available data suggest that holding feeds after transfusion may reduce the risk of NEC.

Avoid prolonged antibiotic use — Observational evidence suggests that infants who receive prolonged antibiotic therapy are at increased risk for NEC [33-35]. In a meta-analysis of 13 studies (7,901 infants), empirical antibiotic therapy for five or more days was associated with increased risk of NEC (adjusted RR 1.51, 95% CI 1.22-1.87) [35]. (See "Neonatal necrotizing enterocolitis: Pathology and pathogenesis", section on 'Microbial dysbiosis'.)

These findings suggest that antibiotic stewardship aimed at reducing unnecessary antibiotic therapy and limiting the duration of empiric antibiotic therapy in infants with sterile cultures may reduce the risk of NEC. (See "Treatment and prevention of bacterial sepsis in preterm infants <34 weeks gestation", section on 'Duration and response to therapy'.)

Avoid gastric acid suppression — Gastric acidity may play a role in preventing the cascade of infectious and inflammatory events leading to NEC. In the neonate, the high-acid environment tightly regulates the amount of pathogenic microbes that colonizes the intestine. H2 blockers such as cimetidine, ranitidine, and famotidine suppress gastric acidity. Observational data from large case series have reported that H2 blockers are associated with an increased risk of NEC [36,37]. In addition, exposure to gastric acid inhibitors may also be associated with increased risk of pneumonia and sepsis. As a result, based on this limited data, we suggest that H2 blockers generally be avoided in VLBW infants. (See "Neonatal necrotizing enterocolitis: Pathology and pathogenesis", section on 'Agents that reduce gastric acidity' and "Gastroesophageal reflux in premature infants", section on 'Safety concerns'.)

The association between H2 blockers and NEC was demonstrated by a study from NICHD of 11,072 VLBW infants born from 1998 to 2001 that reported infants with NEC compared with matched controls were more likely to have received H2 antagonists (odds ratio [OR] 1.71, 95% CI 1.34-2.19) [36].

There are little data on proton pump inhibitors (PPIs) on the incidence of NEC, as these agents are seldom used in preterm neonates. However, we suggest that PPIs also be avoided, since they also reduce gastric acid production and would be expected to have a similar impact on the microbial community in the gut. (See "Gastroesophageal reflux in premature infants", section on 'Acid suppression'.)

Prevent severe anemia/Decrease exposure to RBC transfusion — Although data suggest an association between the development of NEC, anemia, and red blood cell (RBC) transfusion, the nature of the association remains unclear. This association is discussed in detail separately. (See "Neonatal necrotizing enterocolitis: Pathology and pathogenesis", section on 'Anemia and red blood cell transfusion'.)

Strategies we use to prevent severe anemia include delayed cord clamping and limiting phlebotomy when possible. We may also try to prevent severe anemia in VLBW infants with erythropoietin and/or red blood cell (RBC) transfusions. These interventions are discussed separately. (See "Labor and delivery: Management of the normal third stage after vaginal birth", section on 'Preterm infants' and "Anemia of prematurity (AOP)", section on 'Erythropoiesis stimulating agents (ESAs)' and "Red blood cell (RBC) transfusions in the neonate".)

Withhold feeds during and after transfusion — Some data suggest that withholding of feeds during transfusion may reduce the risk of NEC, but evidence remains inconclusive on the optimal management of feeding during transfusion [3,38]. For preterm neonates who require RBC transfusion, we suggest proceeding cautiously with feeds in the immediate post-transfusion period to reduce the risk for NEC. At our center, we use a standardized conservative approach and withhold two feedings after starting a blood transfusion. However, the optimal approach is uncertain, and practice varies from center to center.

Since infants in our unit are bolus fed every three hours, transfusions occur in the middle of a nine-hour period without enteral feeds. The main risk associated with this practice is development of hypoglycemia. Ideally, the transfusion is administered through a separate peripheral intravenous (IV) line that allows for simultaneous infusion of a dextrose solution. In patients who do not have a separate IV line, the infant’s glucose level is checked after the first three hours of the transfusion, and if the level is low, IV dextrose is administered (2 mL/kg of 10 percent dextrose). The transfusion is then resumed to completion. (See "Management and outcome of neonatal hypoglycemia", section on 'Preterm infants'.)

The practice of holding feeds during and after transfusion is supported by observational studies that have reported an association with onset of NEC within 48 hours following a blood transfusion among infants receiving feeds during and after transfusion [39-41]. In a meta-analysis of seven observational studies (7492 neonates), the incidence of NEC was substantially lower in neonates who had feeds held after transfusion compared with those who did not (0.7 versus 2.4 percent; RR 0.47, 95% CI 0.28-0.8). While these observational data cannot establish a causal relationship, the consistent temporal relationship has raised concern for causality. However, it is also plausible that the association between transfusion and NEC is mostly due to the adverse effects of severe anemia rather than the transfusion itself. It has been proposed that severe anemia (eg, hematocrit <25 percent), causes intestinal injury to the gut through hypoxic or immunologic mechanisms, and a subsequent blood transfusion may trigger the development of NEC due to additional changes in viscosity, inflammation, and perfusion to the gut. (See "Neonatal necrotizing enterocolitis: Pathology and pathogenesis", section on 'Anemia and red blood cell transfusion'.)

STRATEGIES NOT ROUTINELY USED

Probiotics — Probiotics, defined by the World Health Organization (WHO) as "live micro-organisms which, when administered in adequate amounts, confer a health benefit on the host," are one of the most studied preventive measures for NEC. They have been shown in a number of studies and meta-analyses to be effective in preventing NEC [42-44]. However, there is lack of consensus on the optimal regimen and insufficient data for extremely preterm infants (gestational age <28 weeks), and the level of evidence was of low quality due to marked heterogeneity and potential bias amongst clinical trials resulting in significant unanswered questions [43].

Therefore, based on the available information and unresolved concerns, routine use of probiotics to prevent NEC is not recommended, consistent with a 2021 Clinical Report from the American Academy of Pediatrics (AAP) [45].

Unanswered or ongoing concerns about the routine use of probiotics include:

●Inconsistent data including potential differences in benefit based on gestational age (GA) and birth weight (BW) – In particular, probiotics may not be as effective in extremely low birth weight (ELBW) infants (BW <1000 g) who are most vulnerable to NEC compared with more mature infants. This was illustrated in a 2017 systematic review and meta-analysis of 23 clinical trials, which were rated as moderate to good quality, that concluded that overall, preterm infants who received probiotics compared with controls had lower all-cause mortality (4.9 versus 6.8 percent; relative risk [RR] 0.72, 95% CI 0.57-0.92) and severe NEC (3.9 versus 6.6 percent; 95% 0.43-0.74) [46]. However, there was no difference in NEC-related mortality (1 versus 1.7 percent; RR 0.64, 95% 0.38-1.07). In addition, for ELBW infants there was no difference between groups for all-cause mortality (8.9 versus 14.3 percent; RR 0.78, 95% CI 0.50–1.20) and severe NEC (8.9 versus 10.6 percent; RR 0.86, 95% CI 0.65-1.16). However, several reviews of the literature note the lack of sufficient data for ELBW infants [43,44].

●Lack of an established regimen of optimal strain, dosing, and timing of administration – Results from the above 2017 review failed to answer the important question of the optimal probiotic strains and doses and duration of therapy [46]. As a result, the authors concluded that a standardized regimen for routine use could not be identified based on the available data. A subsequently published observational study from a single United States tertiary center reported routine administration of Lactobacillus rhamnosus GG ATCC 53103 to very low birth weight (VLBW) infants (BW <1500 g) did not decrease the incidence of NEC between infants in the periods pre- and post-implementation of prophylactic probiotic therapy [47], whereas a retrospective study of preterm infants (GA <29 weeks) from the Canadian Neonatal Network suggested that probiotics, with a dominant use of a multiple strain preparation, were associated with a reduction in the risk of NEC [48]. A 2020 network meta-analysis concluded that combination use of probiotic therapy was more effective than a single probiotic strain [44].

●Serious adverse effects include bacteremia from the bacterial probiotic strain or contamination of the probiotic product. This is an uncommon but serious and potentially fatal adverse event, as illustrated by the following case reports:

•A case report of Bifidobacterium longum bacteremia in two preterm infants demonstrated that the strains isolated from each patient were derived from the probiotic preparations given to the patients to prevent NEC [49].

•A fatal case of mucormycosis was reported in a preterm infant who received the probiotic supplement Solgar ABC Dophilus Powder for prevention of NEC [50]. Investigation of the same lot of unopened Solgar ABC Dophilus Powder by the Centers for Disease Control and Prevention (CDC), US Food and Drug Administration (FDA), and Connecticut Departments of Public Health and Consumer Protection revealed contamination with Rhizopus oryzae. (See "Mucormycosis (zygomycosis)".)

●Lack of quality control regulation to ensure consistency and safety of product, which is required for any approved medication – There is confusion regarding the classification of probiotics, and hence their regulation. Are they a dietary supplement, food ingredient, or drug? One unanswered question is if probiotics should undergo the same regulation for its manufacturing and dispensation as any other medication that is used in a high-risk population such as preterm infants. Stringent quality control requires the manufacturer to produce a consistent product with confirmation of the bacterial taxonomy and colony counts. However, in one study, the contents of 15 out of 16 commercially available products differed from the ingredient list [51]. Strict drug regulation could ensure that the patient is receiving the listed species and subspecies and not receiving any harmful contaminants in the reconstituted product prior to administration, as was highlighted by the unfortunate fatal case of mucormycosis. As of 2021, the US Food and Drug Administration (FDA) has not approved any probiotic product as a therapeutic agent for the prevention of NEC, although investigational studies are underway.

Even if there were unequivocal evidence of benefits without undue harm from probiotic use, challenges that need to be addressed before recommending routine use to prevent NEC include the need to establish a standard regimen of optimal strain and dosing and quality control measures that ensure the administration of a safe and consistent product [52-54]. However, despite the concerns and unanswered questions regarding the safe administration of probiotics, probiotic therapy continues to expand worldwide. If probiotic therapy is used, the European Society for Paediatric Gastroenterology Hepatology and Nutrition (ESPGHAN) Committee on Nutrition and the ESPGHAN Working Group for Probiotics and Prebiotics have provided advice on the selection of specific strains and safety issues that need to be addressed prior to administration [55]. One suggested approach to ascertain local benefit from probiotic use and that we have implemented in our institution is to establish a regimen based on the consensus of all clinicians in practicing group, the involvement of a parent/caregiver representative, and ensure consistent supplementation practices [56]. This would specify eligibility criteria (eg, gestational age and birth weight), product to be used, and clinical protocol (ie, when to start, what dosing regimen to use, and when to stop), efforts to prevent contamination, and monitoring for adverse events, such as laboratory ability to culture the probiotic organism. However, this approach does not mitigate against the above concerns regarding the routine use of probiotics to prevent NEC.

Oral immunoglobulin therapy — The rationale for oral immunoglobulin therapy is that it may reduce NEC by inhibiting release of proinflammatory cytokines [57]. However, based upon the available evidence, immunoglobulin therapy should not be used because the available clinical trials have failed to demonstrate a meaningful benefit from this therapy. In a meta-analysis of three trials (1840 neonates) rates of NEC were similar in neonates assigned to prophylactic oral IgG or IgG/IgA therapy compared with control with did not reduce the incidence of definite NEC (4.7 versus 5.5 percent; relative risk [RR] 0.84, 95% CI 0.57-1.25), suspected NEC (RR 0.84, 95% CI 0.49-1.46), need for surgery (RR 0.21, 95% CI 0.02-1.75), or NEC-related mortality was also similar in both groups (1 percent in both groups; RR 1.1, 95% CI 0.47-2.59) [58].

Nutritional supplements — Nutritional supplements derived from human milk (eg, lactoferrin and oligosaccharides) have been investigated as potential strategies to prevent NEC. However, data either are limited or do not support the use of these interventions to prevent NEC.

Lactoferrin — Lactoferrin is the major whey protein in colostrum, breast milk, tears, and saliva. It is an iron-binding glycoprotein and a component of the mammalian innate response to infection. However, we suggest not using lactoferrin for the prevention of NEC based on the available clinical trial data. In a meta-analysis of seven trials (4874 preterm infants), similar rates of NEC Bell stage II and III were seen in infants assigned to lactoferrin and placebo (RR 1.10 ,95% CI 0.86-1.41) [59]. The lack of efficacy of prophylactic lactoferrin was also seen in the primary outcome of late-onset sepsis, as discussed separately. (See "Treatment and prevention of bacterial sepsis in preterm infants <34 weeks gestation", section on 'Potential prophylactic therapy'.)

Human milk oligosaccharides — Another promising but unproven component appears to be human milk oligosaccharides (HMO), long-chain sugars. There are more than 150 such compounds that represent the third most prevalent constituent in human milk [60]. The possibility that HMOs might prevent NEC was raised by studies in animal models [61]. There are limited clinical data. In a multicenter clinical study, analysis of the breast milk given to preterm infants demonstrated that infants who developed NEC had lower levels of a specific class of HMOs called disialyllacto-N-tetraose (DSLNT) than matched controls, although there was no difference between the two groups regarding the overall amount of total HMOs [62]. A systematic review of the use of plant-based synbiotics(combined probiotics and prebiotics) did not find conclusive evidence of their benefits [63]. Further study is required to evaluate the benefits of HMOs on the preterm gut [63].

Other supplements — The amino acids arginine and glutamine have been studied as possible interventions to prevent NEC:

●Arginine – In a systematic review of the literature, parenteral or enteral arginine supplementation appeared to be protective, with a lower rate of NEC in the group that received arginine compared with those given placebo (RR 0.38, 95% CI 0.23-0.64) [64]. There was no difference in adverse effects and neurodevelopmental outcome between the two groups. Nevertheless, data are inadequate to recommend arginine supplementation to prevent NEC due to the small number of patients that have been enrolled in clinical trials. In addition, arginine is provided to infants who receive total parenteral nutrition, which includes the majority of ELBW infants, who are the most vulnerable patients for NEC.

●Glutamine – In a randomized multicenter trial, parenteral glutamine supplementation was found not to be effective in decreasing the rate of NEC in 721 preterm infants with BW ≤1000 g [65].

As a result, glutamine and arginine should not be given routinely to prevent NEC. Further investigations are needed to determine whether either glutamine or arginine is effective in preventing NEC.

Other supplements derived from human milk, including polyunsaturated fatty acids, erythropoietin, epidermal growth factor, and acetylhydrolase, have been shown to exert a protective role against NEC in experimental models but these results need to be confirmed in clinical trials [66-69].

SUMMARY AND RECOMMENDATIONS

●Effective interventions – Effective strategies to reduce the incidence and/or severity of necrotizing enterocolitis (NEC) in preterm neonates include:

•Antenatal corticosteroids – Pregnant individuals who are at high risk for preterm delivery before 34 weeks gestation should receive antenatal corticosteroids, as discussed in detail separately. (See "Antenatal corticosteroid therapy for reduction of neonatal respiratory morbidity and mortality from preterm delivery".)

•Human milk feeding – Human milk compared with intact bovine milk-formula is associated with a lower risk of NEC. Breast milk feeding should be encouraged for this benefit and because of other well-established benefits of breast milk. If mother's milk is unavailable, pasteurized donor human milk should be used. The benefits of human milk feeding and use of donor milk are discussed in greater detail separately. (See "Human milk feeding and fortification of human milk for premature infants".)

•Standardized feeding protocols – Standardized feeding protocols provide a consistent approach to initiating and advancing feeds, and criteria for when to withhold and restart feeds. The approach is discussed in greater detail separately. (See "Approach to enteral nutrition in the premature infant".)

•Avoiding medications that may contribute to NEC – This includes:

-Antibiotic stewardship to avoid prolonged courses of antibiotics whenever possible (see 'Avoid prolonged antibiotic use' above and "Treatment and prevention of bacterial sepsis in preterm infants <34 weeks gestation", section on 'Duration and response to therapy')

-Avoiding acid suppression unless it is clinically indicated (see 'Avoid gastric acid suppression' above and "Gastroesophageal reflux in premature infants", section on 'Acid suppression')

•Holding feeds during and after transfusions – For very low birth weight (VLBW) infants (BW <1500 g) who require red blood cell (RBC) transfusion, we suggest holding feeds during and after the transfusion (Grade 2C). Observational data suggest that infants who continue to receive feeds during and after RBC transfusion are at higher risk of developing NEC compared with those who have feeds held. (See 'Prevent severe anemia/Decrease exposure to RBC transfusion' above.)

●Unproven or ineffective interventions

•We suggest against the routine use of probiotics (Grade 2C). Although clinical trials have demonstrated that some probiotics decrease risk of NEC, a lack of an established regimen of optimum strain and dosing, absent quality control regulation to ensure product consistency and safety, and questions of which preterm infants would benefit need to be addressed prior to recommending these agents for routine clinical use. (See 'Probiotics' above.)

•We suggest against using lactoferrin or oral immunoglobulins (Grade 2B). The available clinical trial data suggest these interventions provide little to no benefit in reducing the incidence of NEC. (See 'Lactoferrin' above and 'Oral immunoglobulin therapy' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Richard J Schanler, MD, who contributed to an earlier version of this topic review.