Accuracy and Safety of the Cytosponge for Assessing Histologic Activity in Eosinophilic Esophagitis: A Two-Center Study

• DOI: 10.1038/ajg.2017.244
• Publication Date: 2017-08-15T12:02:58Z
• Authors: David A Katzka; Thomas C Smyrk; Jeffrey A Alexander; Debra M Geno; RoseMary A Beitia; Audrey O Chang; Nicholas J Shaheen; Rebecca C Fitzgerald; Evan S Dellon
• Keywords: Adult; Male; Surgical Sponges; Mucous Membrane; Biopsy; Statistics as Topic; Patient Acuity; Cell Count; Patient Preference; Eosinophilic Esophagitis; Middle Aged; United States; Specimen Handling; 3. Good health; Eosinophils; Cross-Sectional Studies; Esophagus; Dimensional Measurement Accuracy; Humans; Female; Esophagoscopy; Prospective Studies; Monitoring, Physiologic

OBJECTIVES: Management of eosinophilic esophagitis (EoE) requires repeated endoscopic mucosal sampling to assess disease activity. A less invasive and expensive means of monitoring of EoE is required. The objective of this study was to assess the accuracy, safety, and tolerability of the cytosponge compared to endoscopy and biopsy for histologic assessment of EoE. METHODS: In this prospective two-center cross-sectional study, patients with known EoE underwent cytosponge sampling followed by endoscopy and biopsy. Sample adequacy and eosinophil counts (eos/HPF) were determined for both cytosponge and endoscopic samples. The cytosponge was assessed for diagnostic accuracy, safety, and patient preference as compared to endoscopy. RESULTS: Six patients (7%) failed to swallow the sponge. One hundred and five procedures were successfully performed in 80 patients (66% male, 100% white, 19% stricture). The cytosponge sample was adequate in 102 and the biopsy in 104; 101 procedures had adequate samples by both techniques. Fifty-seven biopsies were graded as active EoE with ≥15 eos/HPF as the gold standard. Eosinophil counts highly correlated between the biopsy and cytosponge (r=0.78, P<0.0001). Using a cutoff of ≤15 eos/HPF for inactive disease, the sensitivity and specificity of the cytosponge was 75% and 86%, respectively. Six patients had active EoE on cytosponge not found on biopsy. For biopsies with inactive EoE, the cytosponge identified 38/44. No complications occurred, and cytosponge endoscopic abrasion scores were low (0.34/4). Patients preferred cytosponge to endoscopy with higher rating scores (7.27 vs. 6.11, P=0.002). CONCLUSIONS: Compared to endoscopy with biopsy, cytosponge provided a minimally invasive, safe, well tolerated, and accurate method to assess EoE histologic activity. (ClinicalTrial.gov number NCT01585103).