The quest to adequately define, diagnose, and manage human disease is recorded as early as 2600 BC as evidenced by the Egyptian Edwin Smith Papyrus.1 This process continues to be iteratively explored and updated as new approaches (e.g., the scientific method and evidence-based medicine) and new information become available with advancing technologies. Disease classifications can serve many purposes. It can be used to diagnose, predict disease risk, optimize therapeutics, assess morbidity and/or mortality, determine quality of life and health care utilization, and predict long-term medical needs. At the same time, ideally, disease classification should not overdiagnose or overtreat. All these elements are important when considering necrotizing enterocolitis (NEC), the infants it afflicts, and the families it impacts.

It is often discussed that Dr. Martin Bell proposed the first “definition” of NEC in 1978.2 However, this is not a definition of the disease but rather, in Dr. Bell’s words, a “clinical staging” of an entity that was already labeled as NEC. How the diagnosis was made is not fully discussed. In 1986, Drs. Walsh and Kliegman expanded Bell’s staging criteria to include additional stages of severity and therapeutic suggestions with true cases defined as the presence of pneumatosis intestinalis or intrahepatic portal venous gas.3 With this, the term “definite NEC” was assigned to stage IIA and greater and became widely used as a diagnostic definition. However, the modified Bell’s staging leaves just enough ambiguity in the certainty of an NEC diagnosis. The persistence of stage 1 in the algorithm raises the question of whether the presence of pneumatosis or intrahepatic air is a required element for diagnosis.4 The lack of high interobserver reliability of detecting pneumatosis on radiograph or ultrasound complicates this matter further.5 Other conditions can exhibit pneumatosis or pneumoperitoneum but are not NEC, notably spontaneous intestinal perforation (SIP) for the latter. Although there have been improvements in separating SIP from NEC, other diagnoses can present with pneumatosis, pneumoperitoneum, or severe gastrointestinal symptoms with bloody stools that can be difficult to distinguish from NEC and include intestinal obstruction, vascular compromise, and protein-induced enterocolitis.6,7 Furthermore, the modified Bell’s criteria does not address (nor claims to) the question that nags us the most, why does an infant get NEC? What is a reliable set of risk factors that are modifiable and are subject to practice change and early therapies to minimize the risk of disease? To resolve these issues, it rests on us to determine a robust definition of disease presence and here we enter a continuous imperfect circular argument.

Log in or create a free account to read this content

Gain free access to this article, as well as selected content from this journal and more on nature.com

Access through your institution

or

Sign in or create an account
Continue with Google
Continue with ORCiD