Introduction
Endoscopy abnormalities of eosinophilic esophagitis (EoE) are currently graded based on the EoE endoscopic reference score (EREFS) ranging 0 to 9. Research has shown that esophageal biopsies are often omitted in Europe in patients reporting dysphagia when the upper endoscopy (EGD) shows no abnormalities [1]. However, a proportion of patients with eosinophilic esophagitis (EoE) may have normal endoscopy (i.e., EREFS=0), making the diagnosis more difficult and increasing the risk of diagnostic delay.
Aims & Methods
We investigated predictors of a diagnosis of EoE when the EGD is unremarkable (i.e., EREFS=0). Consecutive adult patients (≥18 years) with dysphagia who had EREFS=0 at an EGD with six esophageal biopsies were prospectively enrolled. All patients underwent EGD while off any eosinophil-depleting treatment. EoE with EREFS=0 was diagnosed based on the presence of at least 15 eosinophils/high-power field in at least one of six esophageal biopsies. When all six esophageal biopsies showed <15 eosinophils/high-power field, patients were diagnosed with non-EoE dysphagia (NED). Demographics, diagnostic delay (i.e., interval time between symptoms onset and diagnosis), esophageal symptoms, and atopic comorbidities (i.e., allergic rhinitis, asthma, eczema) were collected for all patients. Pearson's Chi-squared test and Fisher's exact test were used for comparisons. Odds ratios (ORs) were calculated to quantify associations between disease states. Significance threshold was set at p<0.05.
Results
Of 133 included patients, 51 (38%) had NED, while 82 (62%) had EoE with EREFS=0. Patients with EoE were more commonly males (67% vs 35%; p<0.001) and younger (37 vs 46 years; p=0.005) compared to patients with NED. Patients with EoE were diagnosed with a significantly longer diagnostic delay than NED (28 vs 4 months; p<0.001). Heartburn, regurgitation, chest pain and dyspepsia had similar prevalence in the two groups. In contrast, patients with EoE had a significantly higher prevalence of atopic comorbidities compared to NED (83% vs 39%; p<0.001).
The presence of rhinitis, diagnostic delay longer than 12 months, any atopic comorbidity, age younger than 37 years, and male sex were associated with an increased risk of EoE with OR of 47.0, 11.6, 7.4, 3.8, and 3.7, respectively (p<0.01 for all).
Conclusion
EoE remains a possible diagnosis in patients reporting dysphagia even when endoscopy is normal. The presence of rhinitis and longer diagnostic delay, the presence of any atopic comorbidity, and younger age are associated with a diagnosis of EoE and are helpful to raise the suspicion of EoE and prompt endoscopists to collect esophageal biopsies.
| Characteristic | NED (Non-EoE dysphagia) with EREFS 0
N = 51 (38%) | EoE with EREFS 0,
N = 82 (62%) | p-value | Odds Ratio |
| Sex (Male/total) | 18 / 51 (35%) | 55 / 82 (67%) | <0.001 | 3.7 (IQR, 1.6 - 8.3)
|
| Age at diagnosis | 46 (40, 58) | 37 (28, 50) | <0.01 | 3.8 (IQR, 1.4 - 10.8)
|
| Diagnostic delay(months) | 4 (1, 10) | 28 (10, 90) | <0.001 | 11.6 (IQR, 3.8 - 40.8)
|
| Atopy | 20 / 51 (39%) | 68 / 82 (83%) | <0.001 | 7.4 (IQR, 3.1 - 18.2)
|
| Rhinitis | 3 / 51 (5.9%) | 62 / 82 (76%) | <0.001 | 47.6 (IQR, 13.2-264.1)
|
References
[1] Tourlamain, G., Garcia-Puig, R., Gutiérrez-Junquera, C., Papadopoulou, A., Roma, E., Kalach, N., Oudshoorn, J., Sokollik, C., Karolewska-Bochenek, K., Oliva, S., Strisciuglio, C., Bauraind, O., Auth, M. K., Thomson, M., Otte, S., Rok, O., Dias, J. A., Tzivinikos, C., Urbonas, V., Kostovski, A., … ESPGHAN EGID Working group (2020). Differences in Management of Eosinophilic Esophagitis in Europe: An Assessment of Current Practice. Journal of pediatric gastroenterology and nutrition, 71(1), 83–90. https://doi.org/10.1097/MPG.0000000000002672