Original Contribution

The American Journal of Gastroenterology (2001) 96, 758–765; doi:10.1111/j.1572-0241.2001.03618.x

Clinical utility of serodiagnostic testing in suspected pediatric inflammatory bowel disease

S.R. Targan is a cofounder and shareholder of Prometheus Laboratory, San Diego, CA.

Marla C Dubinsky MD1, Joshua J Ofman MD, MSHS2,3, Marnina Urman BSc1, Stephan R Targan MD2,1 and Ernest G Seidman MD1

  1. 1Division of Gastroenterology and Nutrition, Department of Pediatrics, University of Montreal, Montreal, Quebec, Canada
  2. 2Division of Gastroenterology, Departments of Medicine and Health Services Research, Cedars Sinai-Medical Center, UCLA School of Medicine, Los Angeles, California, USA
  3. 3Zynx Health Incorporated (a Subsidiary of Cedars-Sinai Health System), Los Angeles, California, USA

Correspondence: Ernest G Seidman, MD, Chief, GI and Nutrition Division, Ste-Justine Hospital, #5734, 3175 Cote-Ste-Catherine Road Montreal, Quebec, Canada H3T 1C5 .

Received 31 March 2000; Accepted 18 September 2000.





Confronted with nonspecific symptoms, accurate screening tests would be useful to clinicians to distinguish between functional childhood disorders and inflammatory bowel disease (IBD), thus avoiding invasive diagnostic testing. Traditional ulcerative colitis–specific perinuclear antineutrophil cytoplasmic antibody (pANCA) and Crohn's disease–specific anti–Saccharomyces cerevisiae antibody (ASCA) serodiagnostic assays have recently been modified, with ELISA cut-off values recalculated to maximize sensitivity. The aim of this study was to determine whether the combination of these serodiagnostic tests could maximize diagnostic accuracy and minimize invasive investigations in pediatric patients presenting with nonspecific symptoms suggestive of IBD.



With investigators blinded to clinical diagnoses, ASCA, ANCA, and pANCA profiles were obtained prospectively from 128 patients undergoing complete diagnostic evaluation for IBD. In phase I, diagnostic accuracy and predictive values of the modified and traditional assays were compared for the IBD (n = 54) and non-IBD groups (n = 74). In phase II, the overall accuracy of a novel sequential diagnostic testing strategy was determined. Additionally, the potential number of invasive investigations avoided with the hypothetical application of this strategy to the cohort was determined.



For phase I, the modified serodiagnostic assay was more sensitive (81 vs 69%), whereas the traditional assay had a higher specificity (96 vs 72%) for IBD (p < 0.05) For phase II, false-positive diagnoses would have been reduced by 81%, yielding an overall sequential testing strategy accuracy of 84%.



The incorporation of sequential noninvasive testing into a diagnostic strategy may avoid unnecessary and costly evaluations and facilitate clinical decision making when the diagnosis of IBD in children is initially uncertain.