The Index of Severity for Eosinophilic Esophagitis (I-SEE, EoE) is a longitudinal, clinically applicable scoring system designed to evaluate the complete extent of EoE symptoms, endoscopic and histologic activity, fibrostenotic progression, and the risk of complications. We aimed to define the relationship between the I-SEE score and EoE features and disease severity at diagnosis, and to assess whether the I-SEE score could predict the loss of response to budesonide orodispersible tablets (BOT).

Patients with EoE undergoing at least one year of BOT maintenance therapy at the dosage of 1 mg/day were enrolled.  The I-SEE score was calculated at the time of EoE diagnosis. According to the score, patients were classified as mild (1–6 points), moderate (7–14 points) or severe (≥15 points). Patients characteristics and disease features, including symptoms, endoscopy and histology, were collected at baseline and after one year of BOT therapy.  Pearson correlation, chi-square test and logistic regression analysis were used where appropriate. A p value < 0.05 was considered statistically significant.

A total of 114 adult EoE patients (81.6% male) with a mean age of 38.7 (SD±12.8) were retrospectively enrolled in the study. At baseline, 27.2%, 54.4% and 18.4% of patients had mild, moderate and severe I-SEE scores, respectively.

At diagnosis, the total I-SEE score positively correlated with diagnostic delay (r² = 0.18, p = 0.043) and with the EREFS score (r² = 0.40, p < 0.001). Moderate/severe I-SEE was significantly associated with food impaction (χ² = 13.6, p=0.001) the presence of stenosis (χ² = 28.4, p < 0.001) and the need for endoscopic dilation (χ² = 20.9, p < 0.001) at diagnosis. Presence of atopic comorbidities at baseline did not correlate with I-SEE severity score.

In the logistic regression model, increasing I-SEE severity was a strong independent predictor of endoscopic dilation (p < 0.01), corresponding to almost ten-fold increase in risk for each severity category (OR = 9.97, 95% CI:2.8-34.4).

In our cohort, 4.9% has not responded to induction treatment and 24.7% lost histological response after 1 year of BOT. The I-SEE score at diagnosis did not correlate with BOT failure and it could not predict loss of response after one year of follow-up.

In our study, the I-SEE score could not predict loss of response to BOT treatment. However, our results suggest that the I-SEE is a valuable tool for classifying EoE patients according to the severity of their symptoms, endoscopic activity and fibrostenotic features, and for predicting the need for endoscopic dilation at diagnosis.