Neonatal Necrotizing Enterocolitis

Summary

Necrotizing enterocolitis (NEC) is a devastating disease of the gastrointestinal tract that affects mostly premature infants. It is the most significant contributor of gastrointestinal morbidity and mortality in preterm infants. NEC was first described over 60 years ago but despite significant research efforts, its pathogenesis remains poorly understood. A general lack of reliable and specific early clinical or laboratory signs affects the ability to promptly and accurately diagnose this disease and initiate treatment. The absence of a clear definition for NEC impedes the ability to study its epidemiology, pathogenesis, and preventive and treatment modalities. Decades have passed without any significant and meaningful additions to treatment options. Currently, a heterogeneous group of pathologic processes is lumped into one diagnosis termed “NEC.” Those entities are similar in that they can lead to the same outcome—intestinal necrosis. The inciting factors and initial changes in these entities, however, are different. This chapter will describe some of these pathologic processes while focusing on a “classic” form of NEC.

Definition of NEC

A clear definition of any given disease provides a common foundation to study its epidemiology, pathogenesis, outcomes, and effectiveness of treatments. The lack of a clear definition for NEC has complicated efforts directed at studying this disease. A primary cause for the inability to clearly define NEC is the fact that it is not simply one disease entity, but rather a spectrum of conditions. These conditions have a similar outcome—necrosis of the intestine. However, the pathophysiology of these conditions is very different. The chain of events that leads to intestinal necrosis in a term infant with intestinal ischemia caused by a congenital heart lesion is very different when compared to a 6-week-old infant who was born at 24 weeks of gestation. Grouping multiple conditions under one umbrella of NEC affects the reliability of statistical data collected for this disease and stalls efforts to develop effective preventive and treatment strategies ( Table 85.1 ).

Another important issue is developing strict, well-defined diagnostic criteria of the disease. Criteria should be specific, sensitive, and accurate and allow clinicians to differentiate between subsets of NEC and other entities such as spontaneous intestinal perforation (SIP).

This chapter will focus on the “classic” form of NEC. This form of NEC peaks in preterm infants at a similar corrected gestational age and is most often accompanied by clinical deterioration and clear signs of intestinal inflammation, including abdominal distension, periumbilical erythema, bloody stools, elevation of nonspecific inflammatory markers, and distinctive radiologic changes such as pneumatosis intestinalis and portal venous gas.

Epidemiology

Incidence of NEC inversely correlates with gestational age at birth with a higher incidence in babies born at lower gestational ages. About 90% of NEC cases occurred in preterm infants. There is also a correlation between gestational age at birth and length of interval between birth and onset of disease: the earlier an infant is born, the more time will pass between birth and onset of NEC. This results in the highest NEC incidence between 28 and 33 weeks of corrected gestational age. In infants whose gestational age ranges from 22-28 weeks, the incidence of NEC in the United States declined from 13% in 2008 to 9% in 2012. However, the Canadian Neonatal Network reported an average NEC prevalence of 5.9% and in the Japanese Neonatal Research Network it was 1.6%. Infants born at a gestational age of up to 32 weeks were included in this review, which could in part explain lower NEC rates reported in Canada and Japan.

About 10% of cases of NEC occur in term infants, but when compared with premature infants, these cases have a strong association with such risk factors as congenital heart disease, gastroschisis, and/or hypoxic-ischemic events. In addition to gestational age, other risk factors for the development of NEC are being investigated. Race seems to affect the risk for development of NEC, with African-American infants being at higher risk compared to Caucasian. Genetic predisposition is evolving as a risk factor for NEC as well.

Clinical Features

Diagnosis

Diagnosis is made using a combination of clinical, laboratory, and radiologic findings with the latter being more important for determining severity of the disease process. NEC was systematically described and staging criteria proposed by Bell et al. in 1978 with modifications of staging made by Kliegman et al. in 1987. Modified Bell's criteria are widely used for data collection and in research. However, there are limitations to this classification. Stage I consists of a combination of highly nonspecific findings. It significantly distorts the incidence of NEC. Pneumatosis intestinalis, a hallmark of Stage II disease, may be hard to find on radiographs. Additionally, presence of heterogeneously appearing intestinal stool content may be misinterpreted as pneumatosis intestinalis ( Fig. 85.1 ). A hallmark of stage III is perforated viscus with free intraperitoneal air, which could be due to spontaneous intestinal perforation or NEC. True diagnosis remains unknown when a primary peritoneal drain is placed and no direct bowel visualization and histologic evaluation is performed. This classification also falls short when NEC progresses without exhibiting pneumatosis intestinalis or portal venous gas. Even with free intraperitoneal air, if only a drain is placed without direct observation by the surgeon, the definitive diagnosis of NEC cannot be made.

It is important to note that available modes of diagnostics for NEC are able to recognize only changes that occur in the late stages of the disease when the intestinal epithelial barrier has already been compromised and the signs of systemic response and deterioration are obvious.

Decreasing platelet counts, elevated I/T ratio, and elevated or decreased neutrophil counts have been used as laboratory signs of NEC with none of them being specific for this disease. Multiple inflammatory and immunologic markers have been studied as potential predictive biomarkers of NEC. Among them are C-reactive protein (CRP), procalcitonin, an array of cytokines and chemokines, platelet activating factor (PAF), and inter alpha inhibitory protein (IaIp) to name a few. So far, none of them has shown significant ability to do so. Markers of intestinal inflammation and injury such as fatty acid binding protein (FABP), fecal calprotectin (FC), and claudins are being investigated as well. Their usefulness as early screening markers for NEC is questionable. The absence of a clear definition of NEC and limited understanding of its initial pathophysiologic changes impede the ability to develop reliable biomarkers of early disease.

Radiologic signs used to diagnose NEC include pneumatosis intestinalis and portal venous gas ( Fig. 85.2 ) , free intraperitoneal air, distended and fixed loops of bowel, and a gasless abdomen. However, studies have shown significant variability in interpretation of abdominal radiographs of infants with suspected NEC.

Abdominal ultrasound (US) may be emerging as an important imaging modality in evaluating infants with suspected NEC. Its advantages over x-ray are real-time imaging (ability to assess for presence of peristalsis), detection of even minimal ascites, accurate detection of intestinal wall thickness, and perfusion.

The above factors lead to uncertainty in diagnosing early stages of the disease and unnecessary exposure of significant numbers of ELBW infants to interruptions of enteral feedings, courses of broad-spectrum antibiotics, and prolonged need for central access. These interventions are associated with an increased incidence of late onset sepsis, prolongation of parenteral nutrition, and hospitalization.