Necrotizing enterocolitis (NEC) reports appeared as early as the first half of the 19th century. It was named and described as a clinical, radiographic, and pathologic entity in the 60s,[1] following an epidemic that occurred in a Children's Hospital in New York City between 1955 and 1966. In 1978, Bell et al. defined a severity and staging system for NEC to support therapeutic discussion, which is still in use. In the 19th ed.ition of the International Classification of Diseases (ICD 9) published in 1987, an ICD-code for NEC was specified.[2],[3]

NEC is the most common, devastating acquired disease of the gastrointestinal tract in very low birth weight infants (VLBW).[2] NEC proved to be the second most expensive morbidity after bronchopulmonary dysplasia (BPD) in VLBW infants.[4] Prematurity and low birth weights are the most critical risk factors. It has been reported to affect 10%–12% of VLBW infants, with a mortality rate of about 15%–30%.[5],[6]

The pathogenesis of NEC has not been clearly explained.[3] NEC most likely represents a complex interaction of factors causing a mucosal injury that occurs with the coincidence of the following pathologic events: immaturity of the gastrointestinal tract, intestinal ischemia, excess protein substrate in the intestinal lumen,[3],[7] and colonization of the intestine by pathologic bacteria. Preterm infants primarily acquired their colonizing gastrointestinal bacteria from the neonatal intensive care environment, rather than their mother's genital tract flora, skin, or breast milk.[8] Considering that most premature infants do not develop NEC, it has been suggested that there may be a genetic predisposition with fewer mucin-producing goblet cell in some infants.[9]

Most infants with Stage 1 or 2 NEC are managed nonoperatively; however, surgical intervention is considered in the event of intestinal perforation or clinical deterioration.[10] From the infants who survive, about 50% will develop long-term complications, and nearly 10% of these infants will develop late gastrointestinal problems. However, the remaining 50% do not have any long-term sequels.[11] The most common complications are intestinal stricture, short bowel syndrome, total parenteral nutrition (TPN)-related cholestasis, neurodevelopmental delay, prolonged hospital stay, and death.[2],[12],[13]

The incidence of NEC has gradually decreased over the past 10 years, partly due to initiatives directed at preventing the onset of NEC.[14] Currently available strategies for primary prevention of NEC include antenatal glucocorticosteroids, cautious feeding strategy, fluid restriction, breastfeeding, and probiotics.[3] Amniotic fluid stem cell therapy is one of the future prevention interventions in the pipeline where stem cell migrate and colonize the damaged intestinal villi and enhance regeneration of the intestinal epithelium.[15] Furthermore, identifying babies who are at a higher risk of developing NEC by using noninvasive indicators, for example, fecal microbiota and some inflammatory protein tests from the buccal epithelium.[16]

Our study is aimed to evaluate maternal and infant risk factors of NEC. Also, to look at the outcome of all infants that developed NEC during the study period.

In this retrospective cohort study, we used the Electronic Neonatal Database, reviewed the medical charts and radiology reports for all preterm infants admitted to our tertiary hospital from January 2011 to December 2018. All cases diagnosed with NEC were collected and classified based on the Modified Bell's criteria, which included suspected, proven, and complicated NEC. The “bowel stricture” was defined based on clinical assessment and barium study, while “short bowel” was defined based on the length of the bowel removed.

All statistical analysis was performed using the Statistical Package for he IBM SPSS 25.0, Armonk, New York, USA.

Our inclusion criteria included all preterm infants born ≤32 weeks, delivered at our tertiary care hospital, and admitted to the newborn intensive care unit (NICU) with a confirmed diagnosis of NEC from January 2011 to December 2018. Our exclusion criteria included infants diagnosed as spontaneous intestinal perforation, anorectal malformations, congenital gastrointestinal surgical conditions (omphalocele and gastroschisis), and other congenital gastrointestinal diseases like Hirschsprung disease.

The number of admitted babies during the study periods is 5393 babies, Number of babies developed NEC 51 babies. Number of babies with NEC full fill the inclusion criteria, 49 babies.

Out of a total of 1265 admitted infants, 51 infants who met the inclusion criteria were enrolled in our study, 2 of them were excluded because one was born in another hospital, and the second was diagnosed as a case of haemophagocytic syndrome. Consequently, only 49 infants full-filled the inclusion criteria [Figure 1].

Neonatal demographic data mentioned in [Table 1] and maternal demographic data mentioned in [Table 2]. Other studied risk factors for NEC include the use of surfactant in 30 (57%) and no surfactant in 21 (43%) of cases, umbilical arterial line in 24 (49%), and no umbilical arterial line in 25 (51%). Hypotension requiring inotropic support appeared in 16 (32.7%) of the cases while in 29 (59.2%) of the cases it did not require inotropic support. Sepsis associated with NEC developed in 30 (61.2%) of the cases, while sepsis was not reported in 19 (38.8%) of the cases. Packed red blood cell transfusion was observed in 21 (42.9%) of NEC cases.

Over half (53%) of our NEC cases needed surgical intervention, such as penrose or laparotomy, while about (47%) required only medical treatment and observation.

A significant number of patients 7 (14.3%) showed cholestasis on TPN.

Post-NEC recovery complications such as bowel stricture occurred in 4 (8.2%) of our cases and short bowel syndrome occurred in 1 (2%) of our cases. Recurrent NEC was observed in 1 (2%) of our cases.

Neonatal morbidity such as BPD was regarded as an association of NEC in 24 (49%) of the cases with P < 0.001 while retinopathy of prematurity was observed in 9 (18.8%) of our cases with P < 0.001.

To our knowledge, this is one of the largest local studies evaluating babies that developed NEC. Forty-nine preterm VLBW neonates had definite or advanced NEC; the prevalence of NEC was 3.9%, which is less than the internationally reported data. Neonatal Research Network has mentioned that the mean prevalence of NEC in VLBW infants was about 7%.[9] Approximately 35% of babies with NEC died in our hospital. Which is similar to the international mortality rate data for NEC (20%–40%).[3],[9],[17],[18]

In our study group around 31 (63.3%) of the affected neonates were born via cesarean section which is; a known risk factor for NEC.[2],[3] In our study, we found preterm infants born to multigravida mothers had more than a 2-fold (71.4%) increased risk to develop NEC when compared to a preterm infant born to a primigravida mother (28.4%) with significant (P = 0.037) which is significant; although maternal parity is not a known risk factor for NEC.

Maternal comorbidities (hypertension, diabetes mellitus, and chorioamnionitis), the use of intrapartum antibiotics, and prolonged rupture of the membrane (>18 h) did not increase the risk of NEC in our study because of under-reporting. The negative association with NEC included maternal preeclampsia, premature rupture of membrane, and urinary tract infection [Table 2].[2]

Babies born to mothers who did not receive antenatal steroids had higher rates of NEC (79.6%) in comparison to (20.4%) in the group that received antenatal steroids as P value is not significant (0.274). This is contrary to expectations as antenatal steroids have a protective effect against NEC.[3] Antenatal glucocorticoid therapy could reduce the mucosal uptake of macromolecules, decrease colonization with aerobic bacteria, reduce translocation and increase the activity of enzymes such as lactase, maltase, sucrase, and Na\K-ATPase, which have been correlated with a reduction in NEC.[3]

Prematurity has adverse relation on NEC [Figure 2], the more preterm the infant, the more likely the infant will develop NEC; (P = 0.007), which is a well-known risk factor.[1],[2],[3] More than half, 28 (57.2%) of the affected babies in our study weighed <1000 g, 14 (28.5%) weighed 1000–1499 g and only 7 (14.2%) weighed >1500 g with a P= 0.05. Decreased birth weight is also known to be associated with NEC [Figure 3].[6],[17]

Positive blood culture was present in half of our cases, with for [Figure 4] most being Gram-positive, while less than half were Gram-negative cultures. In the literature, studies have revealed the association of sepsis with NEC.[9] Previous studies show that the most common organism identified in NEC infants are Gram-negative (Escherichia coli, Klebsiella pneumonia, Proteus, Staphylococcus aureus, Staphylococcus epidermidis, Enterococcus spp, Clostridium perfringeous, and Pseudomonas aeruginous) while in the majority of our cases has staphylococcus followed by Klebsiella.[10]

Eighty-five percent of the babies who developed NEC were on enteral feeding which is similar to the percentage (90%) reported by Huda et al.[9],[17] The average age at which the infants commenced feeding was 5 days and only 7 (14.3%) developed NEC while they were on “nothing by mouth” (NPO). Most of the feeding interventions that neonatologists adopt aim to encourage the rapid achievement of full enteral feedings to improve postnatal growth, but this approach may potentially increase the risk of NEC. However, the provision of a full package of standardized feeding regimens and practice may decrease the frequency of NEC as adopted and followed in our hospital.[16] Seventy-eight percent reduction in the staging of NEC was observed with the use of a standardized feeding regimen.[13] When feeding was delayed beyond 4 days after birth, it did not cause a reduction in the occurrence of NEC.[12]

About half (47%) of the babies develop NEC while they were on breast milk, about 29% on formula feed and 14.3% were NPO. Breastfeeding has a well-known protective effect on NEC,[14],[18],[19] perhaps due to the bioactive components in a mother's milk that functions synergistically to provide multiple levels of protection from NEC, including immunomodulatory, anti-infective, antioxidant, growth-promoting, and gut colonizing effects. The concentration of bioactive components is highest in colostrum and transitional milk.[14],[17]

Our study showed that the presence of a umbilical artery catheter (UAC) increased the risk of NEC. Fifty-one percent had no UAC and a lesser number of the cases had UAC with a P= 0.027. However, some researchers showed no association between developing NEC and the insertion of UAC in contrast to our study. A study conducted by Rand et al. showed decreased mesenteric blood flow with the use of UAC, and they recommended it to be used carefully in hemodynamically unstable neonates.[20] Similarly, another study showed that the umbilical artery or venous catheter did not seem to increase the risk of NEC.[10]

Hypotension before the development of NEC requiring treatment with inotropes was significant in our study. The literature review showed that hypotension, which required inotropic support is a well-known risk factor for NEC.[5] Surfactant administration was seen in more than half of our NEC patients, which was not significant. Data from studies that investigated the use of surfactant as a risk factor showed huge conflict with some studies showed no association between NEC and the use of surfactant.[4]

Theoretically, red blood cell transfusion is possibly one predisposing factor of NEC; from our research, half of NEC babies had received red blood cell transfusion 48 h before developing NEC. Transfusion-associated NEC (TANEC) is the recently described entity that refers to preterm babies who developed NEC within 48 h of receiving a blood transfusion. TANEC is associated with NEC in over one-fourth of the babies.[3],[7] The risk of TANEC was higher with lower pretransfusion hematocrit.[12] Etiology may relate to an increase proinflammatory cytokine seen after transfusion in neonates, alteration in vascular adaptability after transfusion as well as altered blood flow velocity and reperfusion injury related to the sudden correction of anemia in poorly perfused and oxygenated intestinal tissue.[12] Garg and Sinha performed a meta-analysis of 17 observational studies and did not find an independent association between blood transfusion and NEC.[12]

Abdominal radiographs confirm the diagnosis of NEC and allow following the progression of the disease.[7]

All cases with pneumoperitoneum were diagnosed as Stage 3, which around 59.2%, Stage 2 occurred in 26.5% of total cases, and Stage 1 in 14.3%. Stage 1 is the most challenging stage to be diagnosed because it had a lot of differential diagnoses, for example, viral gastroenteritis, sepsis, and feeding intolerance of prematurity.[2] Pneumoperitoneum is high in our study group may be explained partially by we are not using probiotics yet in our unit. Studies showed probiotics decrease the rate of NEC by around 50%.[8]

Recurrent NEC was not common in our study group, which occurred in only one case during the study period, which is less than the reported percentage[5],[11] Moreover, some literature review also reveals that up to 10% of NEC cases can develop recurrent NEC.[11] Fortunately, the majority of patients with recurrent NEC are successfully treated nonoperatively.[4] Prolong use of TPN increases the chances of TPN-cholestasis, which was significant in our babies, and is known to be an associated complication.[1] In our study group, 7 cases developed TPN complications. One study showed that 42% of surgical NEC developed cholestasis, but this cholestasis does not increase the mortality.[21]

Bowel stricture significantly occurred in a few of our cases. Bowel strictures are known to occur mainly in the left colon in one-third of the cases when an area of the intestine heals with scarring after both operative and nonoperative management.[4],[7] An insignificant number of babies also developed short bowel syndrome. However, from the literature review, short bowel syndrome is the most severe postoperative complication of NEC, occurring in as many as 22%–35% after resection of more than 20 cm.[22] These infants go through a prolonged period of intestinal rehabilitation, depending on the size of the remaining viable gut. The neonatal gut typically grows and adapts over time, but this growth may take up to 2 years. Most infants have a normal gut function at 1–10 years of age.[11] Therefore, 12 of our cases subsequently developed complications because of NEC.

Babies that need surgical intervention including primary peritoneal drainage, laparotomy with resection and enterostomy, resection with primary anastomosis, or proximal diverting jejunostomy. Surgical indications of NEC include pneumoperitoneum, deteriorating infant despite maximum medical treatment, abdominal mass, and signs of persistent intestinal obstruction. Relative indications include increased abdominal tenderness, distention, or persistence portal vein gas.[9] Researchers are still attempting to determine the optimal time for surgical intervention and appropriate techniques to be used.[23] Approximately 46.9% of our study group required surgical intervention while 53.1% were treated by conservative management, which is almost similar to the study results of AlFaleh et al., 27%–63%.[7] Compared to age-matched controls without NEC, infants with medically managed NEC had 22 days longer length of stay and those with surgically managed NEC had 60 days longer length of stay.[4] Among VLBW infants, surgical NEC was associated with significant growth delay and adverse neurodevelopmental outcomes at 18–22 months corrected age compared with no NEC. However, medical NEC does not seem to confer additional risk. Surgical NEC is likely associated with greater severity of the disease.[13] Approximately 50% of neonates who developed NEC have a neurodevelopmental effect, mainly motor developmental delay that seems to be mediated by white matter abnormalities on brain imaging.[4],[10]

From our study, the prevalence of NEC in our NICU was lower, but our mortality rate was similar to those reported. Multigravida mothers have an increased risk of NEC by more than double, more than half of our babies developed advanced NEC, which is double the reported figures found in other studies. Measured to decrease NEC with the use of mother own milk and probiotics is strongly recommended. It is a single-center study; however, it is a sizeable local study identifying maternal, neonatal risk factors and outcome of NEC. The importance of the use of probiotics shown clearly in our study increased advanced NEC but not the rate of NEC. Further local and regional studies are needed to evaluate such serious problems and associated risk factors as well as explore possible preventive measures.

Acknowledgments

The authors would like to thanks the Research Centre, King Fahad Medical City, Riyadh, Saudi Arabia, for providing support in preparing this manuscript.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.