Development of a standardized definition for Hirschsprung's-associated enterocolitis: a Delphi analysis.

Purpose: The reported incidence of Hirschsprung's-associated enterocolitis (HAEC) is extremely variable. A standardized definition would permit comparison of different studies and provide an interpretable outcome measure for future prospective studies in patients with Hirschsprung's disease.

Methods: The Delphi method is a technique for achieving consensus among a panel of experts. A list of 38 potential criteria from the history, physical examination, radiologic studies, and pathologic specimens was made available to pediatric surgeons and gastroenterologists who have contributed to the literature on Hirschsprung's disease. Each expert ranked the diagnostic importance of each item using a Likert scale. In subsequent surveys, the same process was used, but the means and SDs from previous rounds were included as a way of influencing the experts toward consensus. Cronbach's alpha was used after each round to measure variability among the experts. Once consensus was reached, an overall "HAEC score" was developed by assigning a value of 1 or 2 to each item that was considered important by the expert panel. The score was then validated by circulating 10 clinical cases to the panel and asking if each represented HAEC or not.

Results: Twenty-seven experts completed the survey. Cronbach's alpha increased from 0.93 after the first round to 0.97 after the second. Criteria receiving the highest scores were diarrhea, explosive stools, abdominal distension, and radiologic evidence of bowel obstruction or mucosal edema. Eighteen items were included in the score. During the validation process, the score agreed with the experts in 9 of the 10 case scenarios.

Conclusion: The most important clinical diagnostic criteria for HAEC were identified from a larger pool of potential diagnostic items through a consensus approach using the Delphi method. A score was developed and validated and can now be used as a standardized and reproducible outcome measure for future studies in children with Hirschsprung's disease.