Simple tool spots eosinophilic esophagitis without biopsy

Last Updated: 2015-09-07

By Megan Brooks

NEW YORK (Reuters Health) - Four clinical and four endoscopic features can predict eosinophilic esophagitis (EoE) with a high degree of accuracy without the need for biopsy, report gastroenterologists from University of North Carolina, Chapel Hill.

The clinical features are younger age, male sex, and presence of dysphagia and food allergies. The endoscopic features are esophageal rings, furrows, and plaques, and lack of a hiatal hernia.

"This tool can help to gauge clinical likelihood of EoE at the point of care, and in cases where EoE is very unlikely, esophageal biopsies may not need to be obtained," Dr. Evan Dellon, from the Center for Esophageal Diseases and Swallowing at UNC, told Reuters Health by email. "Additionally, the models can help to distinguish EoE from other causes of clinical symptoms in cases where the presentation might not be straightforward."

EoE is often hard to tell from gastroesophageal reflux disease (GERD) and other causes of dysphagia, but the distinction is "critical" because evaluation, treatment and prognosis for the two conditions differ, Dr. Dellon and colleagues noted online August 25 in the American Journal of Gastroenterology.

In a prior retrospective study, they determined that these eight features could independently distinguish EoE from GERD (http://bit.ly/1hLls8P).

Their latest study validates the utility of these factors in prospectively enrolled adults undergoing outpatient upper endoscopy. The cohort included 81 cases of EoE diagnosed according to consensus guidelines and 144 controls who had symptoms of esophageal dysfunction but did not meet criteria for EoE.

Compared with non-EoE cases, EoE cases were younger (38 vs. 52 years; P<0.001), more apt to be male (60% vs. 38%; P=0.001) and white (93% vs. 82%; P=0.03), and nearly all had dysphagia (98%). Both EoE and non-EoE controls had high rates of atopy (69% vs. 58%; P=0.09), but food allergies were more common in those with EoE (43% vs. 15%; P<0.001).

Only 4% of patients with EoE had normal endoscopic exams, the researchers found, with rings (78% vs. 10%), furrows (86% vs. 6%), and plaques (47% vs. 3%) all more common in the EoE patients (P<0.001 for all). In contrast, hiatal hernias were far less common in those with EoE (14% vs. 54%; P<0.001). On histology, EoE cases had 141 eosinophils per high-power field (eso/hpf) on average, compared with 3/hpf for controls (P<0.001).

Confirming their initial observation, the researchers found that the combination of younger age, male sex, dysphagia, and food allergy, plus endoscopic rings, furrows, and plaques, and no hiatal hernia proved highly predictive of EoE.

"The AUC was 0.944, with sensitivity, specificity, and accuracy of 84%, 97%, and 92%," they report. "Similar values were seen after limiting controls to those with only reflux or dysphagia or to those with esophageal eosinophilia not due to EoE."

"Using these predictors at the point-of-care to aid with clinical decision making will avoid the effort and expense of low-yield histological examination for EoE and also provide guidance in cases where differentiating EoE from other conditions is challenging," they write.

The UNC EoE clinical predictor calculator is available here: http://bit.ly/1Xv2Af4.

"I believe that the tool is ready to use to help with clinical decision making," Dr. Dellon told Reuters Health. "This study validated previous preliminary findings in a large prospective cohort of EoE cases and non-EoE controls undergoing endoscopy in an outpatient setting, and the clinical prediction too performed very well."

SOURCE: http://bit.ly/1JI1vXP

Am J Gastroenterol 2015.