Eosinophilic gastrointestinal disorders (EGID) with peripheral eosinophilia: a retrospective review at Mayo Clinic

Hypereosinophilic syndrome (HES) is defined by significant eosinophilia (>1,500 eos/μl), which often leads to end-organ damage/dysfunction. It is unclear if the presence of significant peripheral eosinophilia (>1,500 eos/μl) indicates a more aggressive form of eosinophilic gastrointestinal disorder (EGID).

A database query of the Mayo Clinic Rochester electronic records (1995-2008) was performed using several search terms for eosinophilic gastrointestinal disease, and 161 records were reviewed. Patients under 18 years age, those without Mayo-reviewed pathology specimens, those with eosinophilic esophagitis only, and/or those with evidence of secondary etiologies for GI eosinophilia were excluded. A total of 39 were found to have primary EGID. We compared individuals with biopsy-proven primary EGID based on whether they had significant peripheral eosinophilia (≥1,500 eos/μl) (group A) or not (group B).

Group A tended to have more atopy (A: 12/15; B: 11/24; p = 0.03) and more extensive segmental involvement of the GI tract (p = 0.001). None with available studies had evidence of cardiac (A: 7/15; B: 6/24) or bone marrow (A: 10/15; B: 6/24) involvement. The two thromboembolic events in group A after diagnosis did not translate to significantly greater risk (HR = infinity, p = 0.13; group A vs. B). Doses of initial (A: 40 mg/day; B: 55 mg/day; p = 0.17) and maintenance prednisone (A; 8.75 mg/day; B: 7.5 mg/day; p > 0.90) were similar. Group A was significantly more likely to need maintenance prednisone (77 vs. 8%, p = 0.001), with a median treatment duration of 52 weeks. Recurrence of symptoms (and peripheral eosinophilia) during prednisone taper was common in both groups. Prednisone-sparing agents (hydroxyurea, imatinib mesylate, interferon (IFN)-α2b, anti-interleukin (IL-5) monoclonal antibody) were more commonly used in group A (73 vs. 8%; p < 0.0001).

EGID with peripheral eosinophilia ≥1,500/μl is associated with atopy, greater GI segmental involvement, and uncertain risk of thrombosis. The common use of long-term steroids and variable responsiveness to nonsteroidal agents, particularly in group A, underscores the need for targeted therapies.